@database Amiga_C_List.guide @width 70 @node "Introduction" "Introduction" @wordwrap Amiga-C Mailing List Guide ========================== This guide is intended as an easy reference to the Amiga C mailing list. It contains messages posted to the Amiga C mailing list during the month shown on the main page. To subscribe to the Amiga C mailing list, search for amiga-c at Onelist. http://www.onelist.com This AmigaGuide was created using GuideMaker.rexx © Roger Light 1999 @endnode @node "Main" "Amiga C Mailing List Guide" @{b}Amiga C Mailing List Guide ==========================@{ub} March 2000 @{" Introduction " link Introduction } Introduction to the guide. @{" Search " SYSTEM "RUN <>NIL: Searchguide FINAME"} Search this file (requres SearchGuide) ------------------------------------------------------------------- @{" Linking PPC asm objects to compiled warpup objects (before it drifted onto arrays) " Link "Mail_0"} "Alastair M. Robinson" @{" Re: dos.library Close() won't work! HELP! " Link "Mail_1"} Christian Hattemer @{" Re: dos.library Close() won't work! HELP! " Link "Mail_2"} W1bBle @{" cut scenes " Link "Mail_3"} Steven Dobbs @{" Re: Re: dos.library Close() won't work! HELP! " Link "Mail_4"} tim@ramjam.u-net.com @{" GCC Debug Help Needed " Link "Mail_5"} Matthew Fletcher @{" Copy/move directory " Link "Mail_6"} "Andrzej J. Debicki" @{" 64 Bit Info() " Link "Mail_7"} Colin Wenzel @{" Internet communication " Link "Mail_8"} Charles Barr @{" Re: dos.library Close() won't work! HELP! " Link "Mail_9"} Tim Hanson @{" StormC asm " Link "Mail_10"} Johan Rasten @{" Re: 64 Bit Info() " Link "Mail_11"} Duane McDonnell @{" Re: dos.library Close() won't work! HELP! " Link "Mail_12"} Christian Hattemer @{" Re: 64 Bit Info() " Link "Mail_13"} Duane McDonnell @{" Re: 64 Bit Info() " Link "Mail_14"} Colin Wenzel @{" Re: Re: dos.library Close() won't work! HELP! " Link "Mail_15"} tim@ramjam.u-net.com @{" Re: Re: dos.library Close() won't work! HELP! " Link "Mail_16"} "Francis G. Loch" @{" Re: Internet communication " Link "Mail_17"} Roger Light @{" Re: GCC Debug Help Needed " Link "Mail_18"} Gavin Kinsey @{" Re: StormC asm " Link "Mail_19"} Allan Odgaard @{" Re: StormC asm " Link "Mail_20"} Johan Rasten @{" Re: dos.library Close() won't work! HELP! " Link "Mail_21"} Christian Hattemer @{" Re: Internet communication " Link "Mail_22"} Lee Atkins @{" Re: dos.library Close() won't work! HELP! " Link "Mail_23"} W1bBle @{" Re: Linking PPC asm objects to compiled warpup objects (before it drifted onto arrays) " Link "Mail_24"} Martyn Capewell @{" Re: Re: Smaller executable sizes " Link "Mail_25"} Duane McDonnell @{" STL... " Link "Mail_26"} Alexander Niven-Jenkins @{" Smaller executable sizes " Link "Mail_27"} Niels de Koning @{" Re: Smaller executable sizes " Link "Mail_28"} Colin Wenzel @{" Re: Smaller executable sizes " Link "Mail_29"} Niels de Koning @{" Re: Smaller executable sizes " Link "Mail_30"} Christian Hattemer @{" Re: Smaller executable sizes " Link "Mail_31"} Allan Odgaard @{" Re: Smaller executable sizes " Link "Mail_32"} Ilkka Lehtoranta @{" 508,338 Search Engines and Directories " Link "Mail_33"} WORLD SUBMISSION CENTER @{" Re: Smaller executable size (again) " Link "Mail_34"} GURER OZEN @{" Smaller executable size (again) " Link "Mail_35"} Niels de Koning @{" Re: Smaller executable size (again) " Link "Mail_36"} Ilkka Lehtoranta @{" OT: Missing Digests - a plea " Link "Mail_37"} "Alastair M. Robinson" @{" Re: Smaller exe size " Link "Mail_38"} Colin Wenzel @{" Re: Smaller exe size (again) " Link "Mail_39"} Colin Wenzel @{" PPC bugs with file access " Link "Mail_40"} Steven Dobbs @{" Re: Re: dos.library Close() won't work! HELP! " Link "Mail_41"} "Francis G. Loch" @{" Re: GCC Debug Help Needed " Link "Mail_42"} Matthew Fletcher @{" RE: Internet communication " Link "Mail_43"} roderick_maclean@agilent.com @{" 508,338 Search Engines and Directories " Link "Mail_44"} WORLD SUBMISSION CENTER @{" Re: Smaller executable sizes " Link "Mail_45"} Tim Hanson @{" Re: Missing Digests - a plea " Link "Mail_46"} Tim Hanson @{" Re: Smaller exe size " Link "Mail_47"} Allan Odgaard @{" Re: Missing Digests - a plea " Link "Mail_48"} Allan Odgaard @{" RE: Smaller executable size (again) " Link "Mail_49"} roderick_maclean@agilent.com @{" RE: PPC bugs with file access " Link "Mail_50"} Jesper Svennevid @{" Re: Re: Smaller exe size " Link "Mail_51"} Volker Barthelmann @{" Re: PPC bugs with file access " Link "Mail_52"} Stefan Schulze @{" 508,338 Search Engines and Directories " Link "Mail_53"} "508.338 Search Engines and Directories" @{" RE: 508,338 Search Engines and Directories " Link "Mail_54"} roderick_maclean@agilent.com @{" Re: 508,338 Search Engines and Directories " Link "Mail_55"} Allan Odgaard @{" Chinese " Link "Mail_56"} Boyer Laurent @{" Re: RE: 508,338 Search Engines and Directories " Link "Mail_57"} "Francis G. Loch" @{" Re: Chinese " Link "Mail_58"} "Francis G. Loch" @{" Re: RE: 508,338 Search Engines and Directories " Link "Mail_59"} Andrew Ballard @{" RE: Chinese " Link "Mail_60"} BOYER Laurent @{" RE: Chinese " Link "Mail_61"} "Francis G. Loch" @{" RE: Chinese " Link "Mail_62"} BOYER Laurent @{" BGUI Library " Link "Mail_63"} Shane O'Neill @{" 508,338 Search Engines and Directories " Link "Mail_64"} "508.338 Search Engines and Directories" @{" graphics.library functions under CyberGraphX " Link "Mail_65"} Gabriele Svelto @{" Re: graphics.library functions under CyberGraphX " Link "Mail_66"} Michal Bartczak @{" Does anybody know how to program Linux serial ports? " Link "Mail_67"} Simon John Hall @{" Re: BGUI Library " Link "Mail_68"} "Francis G. Loch" @{" Re: graphics.library functions under CyberGraphX " Link "Mail_69"} Emanuele Cesaroni @{" Re: Does anybody know how to program Linux serial ports? " Link "Mail_70"} Olaf Barthel @{" RE: graphics.library functions under CyberGraphX " Link "Mail_71"} Jesper Svennevid @{" RE: graphics.library functions under CyberGraphX " Link "Mail_72"} Paul Hill @{" pointer array " Link "Mail_73"} Johan Rasten @{" Re: pointer array " Link "Mail_74"} Niels de Koning @{" Re: pointer array " Link "Mail_75"} Andreas Meyer @{" Chinese text editor " Link "Mail_76"} BOYER Laurent @{" Creating ARexx variables " Link "Mail_77"} Richard Drummond @{" RE: Chinese " Link "Mail_78"} "Francis G. Loch" @{" 16bit signed to binary " Link "Mail_79"} Charles Barr @{" Re: 16bit signed to binary " Link "Mail_80"} Tim Hanson @{" Re: pointer array " Link "Mail_81"} Gabriele Svelto @{" Re: pointer array " Link "Mail_82"} Johan Rasten @{" RE: 16bit signed to binary " Link "Mail_83"} Jesper Svennevid @{" Re: 16bit signed to binary " Link "Mail_84"} Robin Cloutman @{" Re: Creating ARexx variables " Link "Mail_85"} amiga-c@tkgbbs.freeserve.co.uk @{" Re: Re: BGUI Library " Link "Mail_86"} Shane O'Neill @{" Re: 16bit signed to binary " Link "Mail_87"} Colin Wenzel @{" Re: Re: pointer array " Link "Mail_88"} Gregory Roberts II @{" RE: Re: 16bit signed to binary " Link "Mail_89"} Jesper Svennevid @{" Reaction " Link "Mail_90"} mike.carter@ukonline.co.uk @{" 14 bit audio " Link "Mail_91"} Paul Hill @{" Re: 14 bit audio " Link "Mail_92"} Paul Hill @{" Re: 14 bit audio " Link "Mail_93"} "Francis G. Loch" @{" Re: 16bit signed to binary " Link "Mail_94"} Robin Cloutman @{" casio databank watches " Link "Mail_95"} Jeffrey Barnes @{" Re: casio databank watches " Link "Mail_96"} Duncan Strand @{" Re: casio databank watches " Link "Mail_97"} Stephan Huebner @{" RE: casio databank watches " Link "Mail_98"} andrewmarkwell@ukonline.co.uk @{" BGUI Library " Link "Mail_99"} "Alastair M. Robinson" @{" Re: What compiler.... " Link "Mail_100"} Colin Wenzel @{" Re: What compiler.... " Link "Mail_101"} Stephan Huebner @{" Re: BGUI Library " Link "Mail_102"} Shane O'Neill @{" Way off topic " Link "Mail_103"} Lee Atkins @{" Dividing ULONGs (was: Re: Way off topic) " Link "Mail_104"} Allan Odgaard @{" Re: What compiler.... " Link "Mail_105"} Colin Wenzel @{" Re: What compiler.... " Link "Mail_106"} Allan Odgaard @{" RE: What compiler.... " Link "Mail_107"} andrewmarkwell@ukonline.co.uk @{" Re: Re: What compiler.... " Link "Mail_108"} Dirk Harlaar @{" StormC and Dev.- CD 2.1 Examples (printer)... " Link "Mail_109"} whose@gmx.de @{" Re: Creating ARexx variables " Link "Mail_110"} Richard Drummond @{" Re: Way off topic " Link "Mail_111"} Duane McDonnell @{" Re: What compiler.... " Link "Mail_112"} Matthew Fletcher @{" RE: FW: Digest Number 312 " Link "Mail_113"} roderick_maclean@agilent.com @{" RE: Re: What compiler.... " Link "Mail_114"} Jesper Svennevid @{" Re: What compiler.... " Link "Mail_115"} Martyn Capewell @{" About future compatibility... " Link "Mail_116"} Gabriele Svelto @{" Re: What compiler.... " Link "Mail_117"} Hynek Schlawack @{" Re: Re: What compiler.... " Link "Mail_118"} Paul Hill @{" RE: Re: What compiler.... " Link "Mail_119"} Jesper Svennevid @{" RE: Re: What compiler.... " Link "Mail_120"} Shane O'Neill @{" Re: What compiler.... " Link "Mail_121"} Colin Wenzel @{" Better editor than se in SAS/C 6.0? " Link "Mail_122"} Neil O'Rourke @{" RE: Better editor than se in SAS/C 6.0? " Link "Mail_123"} "Sealey, M." @{" Re: What compiler.... " Link "Mail_124"} Hynek Schlawack @{" Re: What compiler.... " Link "Mail_125"} Hynek Schlawack @{" Re: Better editor than se in SAS/C 6.0? " Link "Mail_126"} Hynek Schlawack @{" Re: Re: What compiler.... " Link "Mail_127"} Paul Hill @{" Re: Re: What compiler.... " Link "Mail_128"} Paul Hill @{" Re: Re: What compiler.... " Link "Mail_129"} "Andrzej J. Debicki" @{" Re: What compiler.... " Link "Mail_130"} Hynek Schlawack @{" Re: Better editor than se in SAS/C 6.0? " Link "Mail_131"} Matthew Fletcher @{" Re: What compiler.... " Link "Mail_132"} Allan Odgaard @{" RE: Re: What compiler.... " Link "Mail_133"} Matthew Fletcher @{" RE: Re: Better editor than se in SAS/C 6.0? " Link "Mail_134"} "Sealey, M." @{" Re: Re: What compiler.... " Link "Mail_135"} Andrew Ballard @{" Re: Re: What compiler.... " Link "Mail_136"} "Francis G. Loch" @{" Re: What compiler.... " Link "Mail_137"} Hynek Schlawack @{" Re: What compiler.... " Link "Mail_138"} Allan Odgaard @{" Re: What compiler.... " Link "Mail_139"} Allan Odgaard @{" Re: What compiler.... " Link "Mail_140"} Hynek Schlawack @{" Re: Re: What compiler.... " Link "Mail_141"} Bart King @{" Re: Better editor than se in SAS/C 6.0? " Link "Mail_142"} Colin Wenzel @{" RE: Re: What compiler.... " Link "Mail_143"} Volker Barthelmann @{" Re: What compiler / MakeFile " Link "Mail_144"} Colin Wenzel @{" Re: Re: What compiler.... " Link "Mail_145"} Volker Barthelmann @{" Re: Re: What compiler.... " Link "Mail_146"} Laura Vance @{" RE: Re: What compiler.... " Link "Mail_147"} BOYER Laurent @{" FindToolType - Illegal argument... " Link "Mail_148"} Alexander Niven-Jenkins @{" Re: What compiler.... " Link "Mail_149"} Steven Dobbs @{" GRELxxxx gadget borders " Link "Mail_150"} andrewmarkwell@ukonline.co.uk @{" RE: RE: What Compiler " Link "Mail_151"} roderick_maclean@agilent.com @{" Re: What compiler.... " Link "Mail_152"} Hynek Schlawack @{" Re: Re: What compiler.... " Link "Mail_153"} Michal Bartczak @{" Re: What compiler.... " Link "Mail_154"} Hynek Schlawack @{" Re: Re: What compiler.... " Link "Mail_155"} Michal Bartczak @{" Re: What compiler.... " Link "Mail_156"} Hynek Schlawack @{" Re: FindToolType - Illegal argument... " Link "Mail_157"} Christian Hattemer @{" Re: What compiler.... " Link "Mail_158"} Allan Odgaard @{" Re: What compiler.... " Link "Mail_159"} Allan Odgaard @{" Re: What compiler.... " Link "Mail_160"} Allan Odgaard @{" Re: FindToolType - Illegal argument... " Link "Mail_161"} Alexander Niven-Jenkins @{" Re: What compiler.... " Link "Mail_162"} Gabriele Svelto @{" StormC and 68000... " Link "Mail_163"} Alexander Niven-Jenkins @{" Re: STL (Was: What compiler...) " Link "Mail_164"} Alexander Niven-Jenkins @{" Re: What compiler.... " Link "Mail_165"} Hynek Schlawack @{" Re: StormC and 68000... " Link "Mail_166"} Hynek Schlawack @{" Re: Re: What compiler.... " Link "Mail_167"} Shane O'Neill @{" Re: FindToolType - Illegal argument... " Link "Mail_168"} Allan Odgaard @{" Re: FindToolType - Illegal argument... " Link "Mail_169"} Allan Odgaard @{" Re: StormC and 68000... " Link "Mail_170"} Allan Odgaard @{" Re: What compiler.... " Link "Mail_171"} Allan Odgaard @{" #Pragmas & mpega.library " Link "Mail_172"} Mike Carter @{" Re: FindToolType - Illegal argument... " Link "Mail_173"} Christian Hattemer @{" Editors with Syntax highlighting " Link "Mail_174"} Dominic Clifton @{" Re: FindToolType - Illegal argument... " Link "Mail_175"} Christian Hattemer @{" Debugging macro trick " Link "Mail_176"} Tim Hanson @{" A method to define a standard global set of text editing and selecting keyboard shortcuts is needed!! " Link "Mail_177"} Dominic Clifton @{" Re: StormC and 68000... " Link "Mail_178"} Alexander Niven-Jenkins @{" Re: What compiler.... (Specificly, what IDE...) " Link "Mail_179"} Dominic Clifton @{" What PPC compiler (was: What compiler....) " Link "Mail_180"} Dominic Clifton @{" Re: What compiler.... " Link "Mail_181"} Tim Hanson @{" Re: What compiler.... " Link "Mail_182"} Tim Hanson @{" re: What compiler -- Manually talking? " Link "Mail_183"} David Erman @{" Re: Better editor than se in SAS/C 6.0? " Link "Mail_184"} David Erman @{" Re: STL (Was: What compliers...) " Link "Mail_185"} "Nathan J. Young" @{" Re: What compiler -- Manually talking? " Link "Mail_186"} Martyn Capewell @{" Re: What compiler.... (Specificly, what IDE...) " Link "Mail_187"} Nuno Trancoso @{" Re: What compiler -- Manually talking? " Link "Mail_188"} Nuno Trancoso @{" Re: Re: What compiler.... " Link "Mail_189"} Shane O'Neill @{" Re: What compiler.... (Specificly, what IDE...) " Link "Mail_190"} Colin Wenzel @{" Re: StormC and 68000... " Link "Mail_191"} Colin Wenzel @{" Re: Re: What compiler.... (Specificly, what IDE...) " Link "Mail_192"} Shane O'Neill @{" Re: Re: What compiler.... (Specificly, what IDE...) " Link "Mail_193"} Shane O'Neill @{" Re: Better editor than se in SAS/C 6.0? " Link "Mail_194"} Kelly Samel @{" Re: What compiler.... (Specificly, what IDE...) " Link "Mail_195"} Kelly Samel @{" Re: Re: What compiler.... " Link "Mail_196"} Volker Barthelmann @{" Re: Re: What compiler.... " Link "Mail_197"} Volker Barthelmann @{" Re: Debugging macro trick " Link "Mail_198"} Volker Barthelmann @{" stci_d() (was: Re: Better editor than se in SAS/C 6.0?) " Link "Mail_199"} Allan Odgaard @{" Antwort: Re: STL (Was: What compiler...) " Link "Mail_200"} Raimund.Dold@bgt.de @{" How to find the Current Window? " Link "Mail_201"} David Erman @{" Single precision math functions " Link "Mail_202"} Gabriele Svelto @{" Re: What compiler.... " Link "Mail_203"} Hynek Schlawack @{" Re: What compiler.... " Link "Mail_204"} Hynek Schlawack @{" Re: What compiler.... " Link "Mail_205"} Hynek Schlawack @{" Re: stci_d() (was: Re: Better editor than se in SAS/C 6.0?) " Link "Mail_206"} Olaf Barthel @{" RE: #Pragmas & mpega.library " Link "Mail_207"} Jesper Svennevid @{" Re: Re: What compiler.... " Link "Mail_208"} Duane McDonnell @{" RE: What PPC compiler (was: What compiler....) " Link "Mail_209"} Jesper Svennevid @{" Re: How to find the Current Window? " Link "Mail_210"} Colin Wenzel @{" Re: What compiler.... (Specificly, what IDE...) " Link "Mail_211"} Tim Hanson @{" Re: Debugging macro trick " Link "Mail_212"} Tim Hanson @{" Re: What PPC compiler (was: What compiler....) " Link "Mail_213"} Jarmo Laakkonen @{" Re: Re: What compiler.... " Link "Mail_214"} Shane O'Neill @{" ami dev cd probs " Link "Mail_215"} Linus McCabe @{" RE: Re: What PPC compiler (was: What compiler....) " Link "Mail_216"} Jesper Svennevid @{" Re: Better editor than se in SAS/C 6.0? " Link "Mail_217"} Stephen Illingworth @{" Re: stci_d() (was: Re: Better editor than se in SAS/C 6.0?) " Link "Mail_218"} Kelly Samel @{" Re: Re: What compiler.... " Link "Mail_219"} Volker Barthelmann @{" Re: What PPC compiler (was: What compiler....) " Link "Mail_220"} Gavin Kinsey @{" GadTools listviews " Link "Mail_221"} andrewmarkwell@ukonline.co.uk @{" Re: How to find the Current Window? " Link "Mail_222"} David Erman @{" Re: Better editor than se in SAS/C 6.0? " Link "Mail_223"} Kelly Samel @{" Re: How to find the Current Window? " Link "Mail_224"} David Erman @{" Re: Better editor than se in SAS/C 6.0? " Link "Mail_225"} David Erman @{" Re: Re: What compiler.... " Link "Mail_226"} Duane McDonnell @{" Inline " Link "Mail_227"} Bart King @{" BGUI WindowOpen " Link "Mail_228"} Shane O'Neill @{" Re: How to find the Current Window? " Link "Mail_229"} David Erman @{" Re: Inline " Link "Mail_230"} Ilkka Lehtoranta @{" What compiler / MakeFile " Link "Mail_231"} "Alastair M. Robinson" @{" Re: Re: How to find the Current Window? " Link "Mail_232"} Duane McDonnell @{" Re: ami dev cd probs " Link "Mail_233"} Linus McCabe @{" Re: Re: How to find the Current Window? " Link "Mail_234"} Duane McDonnell @{" Re: Re: Inline " Link "Mail_235"} Olaf Barthel @{" Opening libraries... " Link "Mail_236"} Alexander Niven-Jenkins @{" RE: Opening libraries... " Link "Mail_237"} andrewmarkwell@ukonline.co.uk @{" Re: Re: Inline " Link "Mail_238"} Bart King @{" Re: Opening libraries... " Link "Mail_239"} Alexander Niven-Jenkins @{" Re: Opening libraries... " Link "Mail_240"} Hynek Schlawack @{" Re: What compiler.... (Specificly, what IDE...) " Link "Mail_241"} Gabriele Svelto @{" Re: Single precision math functions " Link "Mail_242"} Gabriele Svelto @{" Re: Opening libraries... " Link "Mail_243"} Andrew Bell @{" Re: Opening libraries... " Link "Mail_244"} Steven Dobbs @{" Using MUI with VBCC " Link "Mail_245"} "G. Burdett" @{" watches " Link "Mail_246"} Jeffrey Barnes @{" Re: watches " Link "Mail_247"} Alexander Niven-Jenkins @{" Re: Opening libraries... " Link "Mail_248"} Alexander Niven-Jenkins @{" Re: Using MUI with VBCC " Link "Mail_249"} Hynek Schlawack @{" stormamiga and __stdiowin... " Link "Mail_250"} Alexander Niven-Jenkins @{" Re: Opening libraries... " Link "Mail_251"} Martin McKenzie @{" Re: Opening libraries... " Link "Mail_252"} David Scheibler @{" Port control? " Link "Mail_253"} David Erman @{" Re: Opening libraries... " Link "Mail_254"} Andrew Bell @{" Re: Using MUI with VBCC " Link "Mail_255"} Andrew Bell @{" "Segmentation Fault" sort of OT " Link "Mail_256"} Steven Dobbs @{" Re: Re: Single precision math functions " Link "Mail_257"} Volker Barthelmann @{" Re: Re: Inline " Link "Mail_258"} Duane McDonnell @{" Re: "Segmentation Fault" sort of OT " Link "Mail_259"} Stephen Illingworth @{" Re: "Segmentation Fault" sort of OT " Link "Mail_260"} David Erman @{" Re: Opening libraries... " Link "Mail_261"} Alexander Niven-Jenkins @{" Re: Editors with Syntax highlighting " Link "Mail_262"} Olivier Fabre @{" Re: stormamiga and __stdiowin... " Link "Mail_263"} Allan Odgaard @{" Re: Opening libraries... " Link "Mail_264"} Allan Odgaard @{" Opening Libraries " Link "Mail_265"} "Alastair M. Robinson" @{" RE:Opening Libraries " Link "Mail_266"} roderick_maclean@agilent.com @{" Re: Inline " Link "Mail_267"} Ilkka Lehtoranta @{" RE: Inline " Link "Mail_268"} Jesper Svennevid @{" RE: Re: What PPC compiler (was: What compiler....) " Link "Mail_269"} Jesper Svennevid @{" Re: What PPC compiler (was: What compiler....) " Link "Mail_270"} Fritsch Alexander @{" Re: Re: Inline " Link "Mail_271"} Paul Hill @{" ghostscript docs.. " Link "Mail_272"} Kevin Bewley @{" Re: Single precision math functions " Link "Mail_273"} Gabriele Svelto @{" conio.h? " Link "Mail_274"} Ilkka Lehtoranta @{" Re: ghostscript docs.. " Link "Mail_275"} Gabriele Svelto @{" What is reentrant code? " Link "Mail_276"} Gabriele Svelto @{" Re: What PPC compiler (was: What compiler....) " Link "Mail_277"} Gavin Kinsey @{" Re: Re: Single precision math functions " Link "Mail_278"} Volker Barthelmann @{" RE: conio.h " Link "Mail_279"} roderick_maclean@agilent.com @{" Re: Opening libraries... " Link "Mail_280"} Bart King @{" Re: conio.h? " Link "Mail_281"} Tak Tang @{" Re: ami dev cd probs " Link "Mail_282"} Matthew Fletcher @{" Re: What compiler.... (Specificly, what IDE...) " Link "Mail_283"} Matthew Fletcher @{" Rendering pictures correctly with MUI " Link "Mail_284"} "G. Burdett" @{" (Of Topic) Freeserve DNS for Amiga " Link "Mail_285"} roderick_maclean@agilent.com @{" Re: (Of Topic) Freeserve DNS for Amiga " Link "Mail_286"} David McMinn @{" RE: (Of Topic) Freeserve DNS for Amiga " Link "Mail_287"} "Sealey, M." @{" Whoops " Link "Mail_288"} "G. Burdett" @{" New and some questions, " Link "Mail_289"} Joris Kempen @{" Re: What is reentrant code? " Link "Mail_290"} David McMinn @{" Re: What compiler.... (Specificly, what IDE...) " Link "Mail_291"} Dominic Clifton @{" RE: New and some questions, " Link "Mail_292"} Jesper Svennevid @{" C++ fstream.h problems " Link "Mail_293"} Daithi O'Cuinn @{" Re: conio.h? " Link "Mail_294"} Donald W Millican @{" GCC and AmiTCP includes. " Link "Mail_295"} Krzysztof Wolski @{" Re: New and some questions, " Link "Mail_296"} Krzysztof Wolski @{" Re: New and some questions, " Link "Mail_297"} Joris Kempen @{" Re: New and some questions, " Link "Mail_298"} Alexander Niven-Jenkins @{" Re: New and some questions, " Link "Mail_299"} Allan Odgaard @{" Re: What is reentrant code? " Link "Mail_300"} Allan Odgaard @{" Re: Rendering pictures correctly with MUI " Link "Mail_301"} Allan Odgaard @{" Re: New and some questions, " Link "Mail_302"} Allan Odgaard @{" Re: Re: Single precision math functions " Link "Mail_303"} Duane McDonnell @{" Re: Rendering pictures correctly with MUI " Link "Mail_304"} "G. Burdett" @{" Re: New and some questions, " Link "Mail_305"} Colin Wenzel @{" Re: What is reentrant code? " Link "Mail_306"} Colin Wenzel @{" Re: Rendering pictures correctly with MUI " Link "Mail_307"} Allan Odgaard @{" Re: Opening libraries... " Link "Mail_308"} Allan Odgaard @{" Re: Rendering pictures correctly with MUI " Link "Mail_309"} "G. Burdett" @{" Re: Waiting Task (was: New and some questions) " Link "Mail_310"} Fritsch Alexander @{" Re: Implicit mem allocation (was: Inline) " Link "Mail_311"} Allan Odgaard @{" Re: MUI & palette (Was: Whoops) " Link "Mail_312"} Ilkka Lehtoranta @{" Re: Rendering pictures correctly with MUI " Link "Mail_313"} Allan Odgaard @{" Re: FindToolType - Illegal argument... " Link "Mail_314"} Allan Odgaard @{" Re: What compiler.... " Link "Mail_315"} Allan Odgaard @{" Re: About future compatibility... " Link "Mail_316"} Allan Odgaard @{" Re: StormC and 68000... " Link "Mail_317"} Allan Odgaard @{" Re: Re: Opening libraries... " Link "Mail_318"} Bart King @{" Re: Inline " Link "Mail_319"} Bart King @{" Re: MUI & palette (Was: Whoops) " Link "Mail_320"} "G. Burdett" @{" Re: StormC and 68000... " Link "Mail_321"} Colin Wenzel @{" Check this " Link "Mail_322"} BOYER Laurent @{" Watch out BOYER Laurent " Link "Mail_323"} "Andrzej J. Debicki" @{" RE: Re: Implicit mem allocation (was: Inline) " Link "Mail_324"} Jesper Svennevid @{" RE: Inline " Link "Mail_325"} Jesper Svennevid @{" RE: Watch out BOYER Laurent " Link "Mail_326"} mike.carter@ukonline.co.uk @{" RE:Storm C and 68000 " Link "Mail_327"} roderick_maclean@agilent.com @{" RE: Check this " Link "Mail_328"} Jesper Svennevid @{" RE: Watch out BOYER Laurent " Link "Mail_329"} Jesper Svennevid @{" RE: Watch out BOYER Laurent " Link "Mail_330"} "Sealey, M." @{" Source Level Debugging " Link "Mail_331"} Matthew Fletcher @{" Re: What is reentrant code? " Link "Mail_332"} Gabriele Svelto @{" Re: About future compatibility... " Link "Mail_333"} Gabriele Svelto @{" Re: Inline " Link "Mail_334"} Johan Rasten @{" Re: watches " Link "Mail_335"} Johan Rasten @{" Re: Rendering pictures correctly with MUI " Link "Mail_336"} Gabriele Svelto @{" Re: RE:Storm C and 68000 " Link "Mail_337"} Colin Wenzel @{" Re: Check this " Link "Mail_338"} "Allen J. Newton" @{" Re: Check this " Link "Mail_339"} Andrew Ballard @{" Centering in TEXT_KIND gadgets. " Link "Mail_340"} "Alastair M. Robinson" @{" Re: mp3 player protocol " Link "Mail_341"} Paul Hill @{" Re: Watch out BOYER Laurent " Link "Mail_342"} Andrew Bell @{" ppc-amigaos-gcc 2.95.1 archive " Link "Mail_343"} Gavin Kinsey @{" Re: GCC and AmiTCP includes. " Link "Mail_344"} Gavin Kinsey @{" Changing over to vbcc " Link "Mail_345"} Bart King @{" Re: Centering in TEXT_KIND gadgets. " Link "Mail_346"} Duane McDonnell @{" (unknown) " Link "Mail_347"} Rick Pezzimenti @{" Re: Changing over to vbcc " Link "Mail_348"} Volker Barthelmann @{" Reaction/Boopsi strings " Link "Mail_349"} Kelly Samel @{" Re: Changing over to vbcc " Link "Mail_350"} Allan Odgaard @{" Re: Implicit mem allocation (was: Inline) " Link "Mail_351"} Allan Odgaard @{" Re: mp3 player protocol " Link "Mail_352"} Robert Gnaß @{" OT: Amiga Advanced System Programmers Guide - anyone read it? " Link "Mail_353"} David Erman @{" ReadArgs " Link "Mail_354"} Mike Carter @{" Re: ReadArgs " Link "Mail_355"} Paul Hill @{" Re: RE:Storm C and 68000 " Link "Mail_356"} Tim Hanson @{" Re: Re: Changing over to vbcc " Link "Mail_357"} Bart King @{" WBStartup messages " Link "Mail_358"} Bart King @{" Re: mp3 player protocol " Link "Mail_359"} Mike Carter @{" Re: Source Level Debugging " Link "Mail_360"} Volker Barthelmann @{" Re: WBStartup messages " Link "Mail_361"} Andrew Bell @{" Re: ReadArgs " Link "Mail_362"} Duane McDonnell @{" Enforcer hits " Link "Mail_363"} radiat-r aka Sascha Ploss @{" Re: Pragma,Proto etc. was New and ... " Link "Mail_364"} Joris Kempen @{" RE: Enforcer hits " Link "Mail_365"} "Sealey, M." @{" Re: Re: Pragma,Proto etc. was New and ... " Link "Mail_366"} David McMinn @{" RE: ReadArgs " Link "Mail_367"} roderick_maclean@agilent.com @{" Re: Re: Pragma,Proto etc. was New and ... " Link "Mail_368"} Allan Odgaard @{" RE: Enforcer hits " Link "Mail_369"} Jesper Svennevid @{" Re: ReadArgs " Link "Mail_370"} amiga-c@tkgbbs.freeserve.co.uk @{" Re: Enforcer hits " Link "Mail_371"} Colin Wenzel @{" RE: Enforcer hits " Link "Mail_372"} Olaf Barthel @{" Re: Enforcer hits " Link "Mail_373"} Charles Barr @{" Re: Enforcer hits " Link "Mail_374"} Alexander Niven-Jenkins @{" Actively getting Messages " Link "Mail_375"} Joris Kempen @{" Re: Re: Enforcer hits " Link "Mail_376"} meyersrl@songs.sce.com @{" Re: Re: Enforcer hits " Link "Mail_377"} Neil O'Rourke @{" Re: Enforcer hits " Link "Mail_378"} Allan Odgaard @{" Re: ReadArgs " Link "Mail_379"} Duane McDonnell @{" Re: ReadArgs " Link "Mail_380"} Duane McDonnell @{" Re: Enforcer hits " Link "Mail_381"} Colin Wenzel @{" Re: Enforcer hits " Link "Mail_382"} Colin Wenzel @{" Re: Enforcer hits " Link "Mail_383"} Alexander Niven-Jenkins @{" RE: Re: Enforcer hits " Link "Mail_384"} "Sealey, M." @{" Re: Re: Enforcer hits " Link "Mail_385"} Neil O'Rourke @{" Re: Re: Enforcer hits " Link "Mail_386"} tim@ramjam.u-net.com @{" Re: Enforcer hits " Link "Mail_387"} Tim Hanson @{" Re: Actively getting Messages " Link "Mail_388"} Tim Hanson @{" Re: mp3 player protocol " Link "Mail_389"} David Erman @{" AllocMem Was: Re: Enforcer hits " Link "Mail_390"} Steven Dobbs @{" TManager beta-testing " Link "Mail_391"} Jesper Svennevid @{" Re: Actively getting Messages " Link "Mail_392"} Colin Wenzel @{" Re: Enforcer hits " Link "Mail_393"} Andrew Bell @{" Re: Re: Enforcer hits " Link "Mail_394"} tim@ramjam.u-net.com @{" Re: AllocMem Was: Re: Enforcer hits " Link "Mail_395"} Charles Barr @{" Re: AllocMem Was: Re: Enforcer hits " Link "Mail_396"} GURER OZEN @{" Re: AllocMem Was: Re: Enforcer hits " Link "Mail_397"} Tim Hanson @{" Changing colors.. " Link "Mail_398"} Joris Kempen @{" Re: Changing colors.. " Link "Mail_399"} Allan Odgaard @{" Re: AllocMem Was: Re: Enforcer hits " Link "Mail_400"} Allan Odgaard @{" Re: Enforcer hits " Link "Mail_401"} Stephen Illingworth @{" Re: Enforcer hits " Link "Mail_402"} Allan Odgaard @{" Re: Re: WBStartup messages " Link "Mail_403"} Bart King @{" stdout... " Link "Mail_404"} Allan Odgaard @{" Re: Actively getting Messages " Link "Mail_405"} Duane McDonnell @{" Re: Re: Enforcer hits " Link "Mail_406"} Duane McDonnell @{" Re: Re: Enforcer hits " Link "Mail_407"} Volker Barthelmann @{" RE: Re: Enforcer hits " Link "Mail_408"} Duane McDonnell @{" Re: Re: Actively getting Messages " Link "Mail_409"} Duane McDonnell @{" Re: stdout... " Link "Mail_410"} Andrew Bell @{" Re: Re: stdout... " Link "Mail_411"} Allan Odgaard @{" Re: Actively getting Messages " Link "Mail_412"} Tim Hanson @{" Re: stdout... " Link "Mail_413"} BlackMael @{" Re: TManager beta-testing " Link "Mail_414"} Mike Carter @{" Re: stdout... " Link "Mail_415"} Allan Odgaard @{" RE: Re: TManager beta-testing " Link "Mail_416"} Jesper Svennevid @{" Test message, Please delete. " Link "Mail_417"} Andrew Ballard @{" Duplicate Emails " Link "Mail_418"} mike.carter@ukonline.co.uk @{" ObtainPen for my colors " Link "Mail_419"} Joris Kempen @{" RE: ObtainPen for my colors " Link "Mail_420"} andrewmarkwell@ukonline.co.uk @{" RE: Pragma,Proto etc. was New and ... " Link "Mail_421"} andrewmarkwell@ukonline.co.uk @{" RE: ObtainPen for my colors " Link "Mail_422"} David McMinn @{" Re: Actively getting Messages " Link "Mail_423"} Ilkka Lehtoranta @{" RE: ObtainPen for my colors " Link "Mail_424"} andrewmarkwell@ukonline.co.uk @{" RE: ObtainPen for my colors " Link "Mail_425"} David McMinn @{" Re: ObtainPen for my colors " Link "Mail_426"} Allan Odgaard @{" Re: Pragma,Proto etc. was New and ... " Link "Mail_427"} Allan Odgaard @{" Re: Inline " Link "Mail_428"} Allan Odgaard @{" Optimising C in ASM " Link "Mail_429"} Lee Atkins @{" SAS-C -> StormC " Link "Mail_430"} Mike Carter @{" Re: Re: Actively getting Messages " Link "Mail_431"} Duane McDonnell @{" Re: Re: Pragma,Proto etc. was New and ... " Link "Mail_432"} Paul Hill @{" Re: SAS-C -> StormC " Link "Mail_433"} Bart King @{" endian code donations? " Link "Mail_434"} Steven Dobbs @{" Re: endian code donations? " Link "Mail_435"} David McMinn @{" Re: ObtainPen for my colors " Link "Mail_436"} Joris Kempen @{" Re: Re: ObtainPen for my colors " Link "Mail_437"} David McMinn @{" Re: Re: ObtainPen for my colors " Link "Mail_438"} Paul Hill @{" Re: ObtainPen for my colors " Link "Mail_439"} Jesper Svennevid @{" Re: Pragma,Proto etc. was New and ... " Link "Mail_440"} Allan Odgaard @{" Re: endian code donations? " Link "Mail_441"} Allan Odgaard @{" Re: endian code donations? " Link "Mail_442"} Steven Dobbs @{" RE: Re: ObtainPen for my colors " Link "Mail_443"} Jesper Svennevid @{" Re: SAS-C -> StormC " Link "Mail_444"} Allan Odgaard @{" Re: ObtainPen for my colors " Link "Mail_445"} Joris Kempen @{" Re: ObtainPen for my colors " Link "Mail_446"} Joris Kempen @{" RE: Re: ObtainPen for my colors " Link "Mail_447"} Jesper Svennevid @{" Re: Re: endian code donations? " Link "Mail_448"} Damir Arh @{" Re: ObtainPen for my colors " Link "Mail_449"} Allan Odgaard @{" Re: ObtainPen for my colors " Link "Mail_450"} Allan Odgaard @{" Buffered displays " Link "Mail_451"} Bart King @{" RE: ObtainPen for my colors " Link "Mail_452"} andrewmarkwell@ukonline.co.uk @{" Odp: 14 bit audio " Link "Mail_453"} Michal Bartczak @ENDNODE @NODE "Mail_0" " Linking PPC asm objects to compiled warpup objects (before it drifted onto arrays) " From: "Alastair M. Robinson" Date: 02-29-00 22:33:12 Subject: @{b}Linking PPC asm objects to compiled warpup objects (before it drifted onto arrays)@{ub} From: "Alastair M. Robinson" Hi Gabriele, [Apologies to anyone who objects to reading about assembler on a C-list!] >> The powerpc.library has a RunPPC function which can be used from 68k >> code. If you haven't got the WarpOS documentation, get it. You could >> also have a peep at the VBCC-WarpOS startup-code source. > I should take a look at the docs indeed but I hadn't time lately I know the feeling! >> I would recommend that you write your time-critical function in C (or >> maybe some simpler toy functions), compile it with vc -S +warpos and >> study the PPC .asm file which results. > I'm trying but it's quite hard, PPC asm looks a lot more complex than > 68k asm. Yeah, it's more complicated at first sight. The most important concept is that the *first* operand to each instruction is the destination. Once you're comfortable with that, things make a little more sense. Persevere with reading the code produced for toy C functions! >> C code compiled for PPC is usually much more efficient than compiled >> 68k code, and hand-optimised assembler won't give anything like the >> speedup you're used to from 68k programming. Since this is the Amiga-C >> list, I recommend you do as much as possible in C! > Is this due to pipelining in the PPC processor? Id est: the compiler > knows how to use it at the best? Anyhow I always write first all Up to a point yes, but it's more to do with the PPC having 32 registers to play with. When a compiler runs out of registers it starts parking data on the stack, which of course takes time. Also, compilers frequently create less-than-intelligent code which shuffles values all over the machine's registers. Removing this kind of redundancy gives a large speedup on 68k, but a fast PPC can do dozens of moves between registers in the time taken to do one memory access, so hand-optimising hardly makes any difference, even if the compiler knows nothing about pipelining. > functions in C and only when I get them working correctly I move them to > asm. Definitely the best strategy. > Er... which is the equivalent of the user stack pointer? r13 or r1? I /think/ it's r1 - again VBCC's generated code should provide some clues! All the best, -- Alastair M. Robinson, email:blackfive@fakenhamweb.co.uk Amiga 4000/030, 18Meg RAM, 4.3Gig HD, Mustek ScanExpress 6000SP C, Assembler (68k, PIC16C84, ATMEL AVR), PostScript, ARexx, AMOSPro(!), AmigaDOS Unless you are very rich and very eccentric, you will not enjoy the luxury of a computer in your own home. -- Edward Yourdon, 1975. ------------------------------------------------------------------------ DON'T HATE YOUR RATE! Get a NextCard Visa, in 30 seconds! Get rates as low as 0.0% Intro or 9.9% Fixed APR and no hidden fees. Apply NOW! http://click.egroups.com/1/2120/0/_/451227/_/952128868/ ------------------------------------------------------------------------ -> Spread the URL: http://www.onelist.com/subscribe/amiga-c <- @ENDNODE @NODE "Mail_1" " Re: dos.library Close() won't work! HELP! " From: Christian Hattemer Date: 03-01-00 00:47:19 Subject: @{b}Re: dos.library Close() won't work! HELP!@{ub} From: Christian Hattemer On 29-Feb-00 tim@ramjam.u-net.com wrote: > The files in .deldir do not occupy any space - it's not like > Windows where you have to empty the trash. The goal is the remove all files from the .deldir, may be useful if you don't use the computer alone. As this isn't possible the only way is to create and delete so much files that the .deldir only contains these empty files. However the data of the files is still on the disk after that, only the references in the filesystem have been removed. It may still be possible to restore them with a undelete program which scans the whole disk, but I don't know anyone doing this for PFS. If there is one it's certainly not that easy as looking in the .deldir. > If you want to overwrite the contents of the files in .deldir > (say with 0s) then you need to do that before you actually delete > them - write a copy of the delete command which overwrites the > file then deletes it. This won't work. You'll end up with two copies of the file in the .deldir. The first one is the original file which is moved there when you overwrite it with zeros. When you delete the file with the zeros then it's also moved to the .deldir. Bye, Chris -- __ /// __ /// Amiga 4000T 040/40 Team *AMIGA* \\\/// \XX/ http://www.riednet.wh.tu-darmstadt.de/~chris/ "Der nächste bitte", sagte der Arzt. Er meinte den Schein. ------------------------------------------------------------------------ Show your style! Choose from 6 great card designs when you apply for Capital One's 9.9% Fixed APR Visa Platinum. http://click.egroups.com/1/1894/0/_/451227/_/951868699/ ------------------------------------------------------------------------ -> Spread the URL: http://www.onelist.com/subscribe/amiga-c <- @ENDNODE @NODE "Mail_2" " Re: dos.library Close() won't work! HELP! " From: W1bBle Date: 03-01-00 02:12:58 Subject: @{b}Re: dos.library Close() won't work! HELP!@{ub} From: W1bBle Hi all, Just noticed this thread get resurrected again! :) Cool. If you go to aminet there's a program I wrote (with a little help from some folks on this list - cheers guys!) that does just what has been suggested - removing all the entries in the .DELDIR that another party (your siblings, other half, NSA, GCHQ - whoever can get you in the most trouble! :-P ) may be interested in. The source code is in the archive for those who are interested.... >> The files in .deldir do not occupy any space - it's not like >> Windows where you have to empty the trash. > > The goal is the remove all files from the .deldir, may be useful if you > don't use the computer alone. As this isn't possible the only way is to > create and delete so much files that the .deldir only contains these empty > files. However the data of the files is still on the disk after that, only > the references in the filesystem have been removed. It may still be > possible to restore them with a undelete program which scans the whole > disk, but I don't know anyone doing this for PFS. If there is one it's > certainly not that easy as looking in the .deldir. Spot on. I couldn't have put it better. A program that repeatedly creates and deletes files is only going to foil a casual spy. Anyone /really/ interested in getting at the files whose entries you've just removed from the .DELDIR would (probably?) be forced to scan your entire hard drive, looking for data structures the file system has created but no longer keeps user viewable references to. So something like DiskSalv for PFS I suppose (if there existed such a thing). >> If you want to overwrite the contents of the files in .deldir >> (say with 0s) then you need to do that before you actually delete >> them - write a copy of the delete command which overwrites the >> file then deletes it. > > This won't work. You'll end up with two copies of the file in the .deldir. > The first one is the original file which is moved there when you overwrite > it with zeros. When you delete the file with the zeros then it's also > moved to the .deldir. Yup :) Like I said (one more shameless plug!) check out the source in PurgeDELDIR.lha in disk/misc. It's fairly well commented and everything! :) I've been using it repeatedly for the past week (ie. since it was completed) and it works fine. Use it on the partitions you've been working on before shutdown, or before you hand the computer over to someone else, for a bit more protection from the "idly curious" :) Byee Giles -- <------------------------------------------------> | Giles Burdett __ ___ | | IRC nick = W1bBle /// (. .) | | /// ( v ) | | A1200T 200MHz PPC __ /// // \\ | | OS 3.1 / Linux-m68k \\\/// .X # X. | | \XX/ XX___XX | <------------------------------------------------> ------------------------------------------------------------------------ Registering a domain name is quick and easy. http://click.egroups.com/1/1611/0/_/451227/_/951877389/ ------------------------------------------------------------------------ -> Spread the URL: http://www.onelist.com/subscribe/amiga-c <- @ENDNODE @NODE "Mail_3" " cut scenes " From: Steven Dobbs Date: 03-01-00 03:38:24 Subject: @{b}cut scenes@{ub} From: Steven Dobbs Hello, I'm writing a mission based 3d space game - I will be doing cut scenes but unfortunately I dont know how to get them played. Is there a way I could use some freeware player that is executed from a program, And on completion of the animation, my main program would resume? if so could somebody please explain how I do that. Regards -- Steven Dobbs ------------------------------------------------------------------------ Tired of missing calls while online? Now you can surf the Internet without worrying about missing important calls! CallWave's FREE Internet Answering Machine lets you hear who's calling while online. http://click.egroups.com/1/1927/0/_/451227/_/951878370/ ------------------------------------------------------------------------ -> Spread the URL: http://www.onelist.com/subscribe/amiga-c <- @ENDNODE @NODE "Mail_4" " Re: Re: dos.library Close() won't work! HELP! " From: tim@ramjam.u-net.com Date: 03-01-00 11:04:43 Subject: @{b}Re: Re: dos.library Close() won't work! HELP!@{ub} From: tim@ramjam.u-net.com Hi Christian > > If you want to overwrite the contents of the files in .deldir > > (say with 0s) then you need to do that before you actually delete > > them - write a copy of the delete command which overwrites the > > file then deletes it. > > This won't work. You'll end up with two copies of the file in the .deldir. > The first one is the original file which is moved there when you overwrite > it with zeros. When you delete the file with the zeros then it's also moved > to the .deldir. Only if you open the file for write (i.e. creating a new file with the same name as the old one). If you open the file for *UPDATE* then you can overwrite the actual contents of that file. No .deldir entry is created, because no file is deleted by the open. Tim -- ========================================================================== Ramjam Consultants Ltd, 28 Queensway, Caversham, Reading, RG4 6SQ, England e-mail: support@ramjam.u-net.com http://www.ramjam.u-net.com/ telephone: +44 (0)118 946 5940 fax: +44 (0)118 946 5941 ------------------------------------------------------------------------ Registering a domain name is quick and easy. http://click.egroups.com/1/1611/0/_/451227/_/951905090/ ------------------------------------------------------------------------ -> Spread the URL: http://www.onelist.com/subscribe/amiga-c <- @ENDNODE @NODE "Mail_5" " GCC Debug Help Needed " From: Matthew Fletcher Date: 03-01-00 12:08:49 Subject: @{b}GCC Debug Help Needed@{ub} From: "Matthew Fletcher" Hi, smart people, i have asked similar questions before but i now really need help. I need to get barfly working with gcc, i am aware that barfly supports the stabs debug format but i have never managed to get barfly to open the code in source mode. I use gcc 1.cpp -o 1.exe -g -gstabs, but barfly says 'ea not in hunk', i know ixemuls trap handeler can cause problems but libnix wont work either, its says 'Hunk has no debug information', the docs say to use gerlib which is long dead. If the problem is the lack of amigaos hunk support can anyone mail me a specs file setup for a diffrent linker/assembler that they know works. regards, -- amimjf@connectfree.co.uk ------------------------------------------------------------------------ Registering a domain name is quick and easy. http://click.egroups.com/1/1611/0/_/451227/_/951908938/ ------------------------------------------------------------------------ -> Spread the URL: http://www.onelist.com/subscribe/amiga-c <- @ENDNODE @NODE "Mail_6" " Copy/move directory " From: "Andrzej J. Debicki" Date: 03-01-00 12:23:57 Subject: @{b}Copy/move directory@{ub} From: "Andrzej J. Debicki" Hello, Can someone provide me with sample source code doing copy/move directory? But not using copy shell command but functions and structures from AmigaOS. TIA -- _ # Andrzej J. Debicki # A12ooT 66MB \\ _ # PANS - Polish Amiga News Squad # 6o3e/o6o \\// # denbi@power-amiga.prv.pl # BVisionPPC \/ # http://www.power-amiga.prv.pl # AmigaOS 3.5 ----------------------------------------------------------- ------------------------------------------------------------------------ Registering a domain name is quick and easy. http://click.egroups.com/1/1611/0/_/451227/_/951909836/ ------------------------------------------------------------------------ -> Spread the URL: http://www.onelist.com/subscribe/amiga-c <- @ENDNODE @NODE "Mail_7" " 64 Bit Info() " From: Colin Wenzel Date: 03-01-00 15:00:15 Subject: @{b}64 Bit Info()@{ub} From: Colin Wenzel Just wondering if anyone has a good routine to support 64 bit values from Info(). I have a "customised" dosinfo() I use to return the bytes free on a volume, but of course it now breaks on big drives > 4 gig. here is the important bit: //---- 'info' is a pointer to an aligned InfoData structure. BytesFree = ((info->id_NumBlocks) - (info->id_NumBlocksUsed)) * (info->id_BytesPerBlock); /---------------------------------------------- Now here's the clencher, it has to work on version 37 so you can forget the UMult64() in utility library.... I would like to just return 2 'longs' , one containing the "Gigabytes" bit & the other, the Bytes left over part, so I can just use RawDoFormat like this: sprintf( tempbytebuf, "%09ld", BytesFree ); printf( "%ld.%.3s GB ", GigabytesFree, tempbytebuf ); This gives an output like this: " 12.543 GB " I used a routine (below), but it is slow & clumsy & I just don't like it..... Here it is anyway.... REG ULONG bpb, blocks; #define ONEGIG 1000000000L blocks = (info->id_NumBlocks) - (info->id_NumBlocksUsed); Bytesfree = 0L; GigaBytesFree = 0L; for(bpb = info->id_BytesPerBlock; bpb; bpb --){ BytesFree += blocks; if( BytesFree >= ONEGIG ){ GigaBytesFree ++; BytesFree -= ONEGIG; } } This works, but requires 512, 1024 (or whatever blocksize) number of itterations to execute... I really need another perspective please..... PS: I know ONEGIG is (1L<<30) but it doesn't work in this configuration because of the carry... ------------------------------------- Colin Wenzel. Australia. EMAIL: mailto:colstv@hotkey.net.au URL: http://www.hotkey.net.au/~colstv/ ICQ: 17608330 AMIGA: 4000T, 060/50, 150Mb RAM, O.S 3.5 --------------------------------------- ------------------------------------------------------------------------ Registering a domain name is quick and easy. http://click.egroups.com/1/1611/0/_/451227/_/951919278/ ------------------------------------------------------------------------ -> Spread the URL: http://www.onelist.com/subscribe/amiga-c <- @ENDNODE @NODE "Mail_8" " Internet communication " From: Charles Barr Date: 03-01-00 19:05:46 Subject: @{b}Internet communication@{ub} From: Charles Barr Hello, I want to add a feature to my program, which when it is started contacts the site and finds out if there is a new version of itself. Anyone any ideas on how i could do this? Cheers -- Among the detectives and patrol officers crowded over the body of Latonya Kim Wallace there is no easy banter, no course exchange of cop humor or time-worn indifference. -- David Simon ------------------------------------------------------------------------ Registering a domain name is quick and easy. http://click.egroups.com/1/1611/0/_/451227/_/951933829/ ------------------------------------------------------------------------ -> Spread the URL: http://www.onelist.com/subscribe/amiga-c <- @ENDNODE @NODE "Mail_9" " Re: dos.library Close() won't work! HELP! " From: Tim Hanson Date: 03-01-00 19:52:35 Subject: @{b}Re: dos.library Close() won't work! HELP!@{ub} From: Tim Hanson Hi Christian Christian Hattemer wrote this on 01-Mar-00 at 00:47:19 > From: Christian Hattemer > > On 29-Feb-00 tim@ramjam.u-net.com wrote: > >> The files in .deldir do not occupy any space - it's not like >> Windows where you have to empty the trash. > > The goal is the remove all files from the .deldir, may be useful if you > don't use the computer alone. > As this isn't possible the only way is to create and delete so much files > that the .deldir only contains these empty files. > However the data of the files is still on the disk after that, only the > references in the filesystem have been removed. It may still be possible to > restore them with a undelete program which scans the whole disk, but I don't > know anyone doing this for PFS. > If there is one it's certainly not that easy as looking in the .deldir. I`ve got DsikSalv V4 which recognises AFS disks but will not do undelete. However if you want to be realy safe, write a delete replacement that creates a file with the same name of exactly the same size then delete that. I think that should wipe the original data completely. -- Tim Hanson, by the River,Canterbury, Kent. Amiga A1200T,`040/25,603e/160,32MB Fast,BVisionPPC, 8.4GB IDE HD,ZIP SCSI,RENO x2CD SCSI,Pace 56K Don`t waste your time by visiting http://www.sinister67.freeserve.co.uk ------------------------------------------------------------------------ Registering a domain name is quick and easy. http://click.egroups.com/1/1611/0/_/451227/_/951946225/ ------------------------------------------------------------------------ -> Spread the URL: http://www.onelist.com/subscribe/amiga-c <- @ENDNODE @NODE "Mail_10" " StormC asm " From: Johan Rasten Date: 03-01-00 22:13:14 Subject: @{b}StormC asm@{ub} From: Johan Rasten Is there a better way to send parameters to asm functions from C/C++ than using globals? I'm also interested in retrieving the results of the func. Can I inline asm in my C++ progs? Please help! I bought the OS dev2.1, but there's no manual for StormC ;,( Live long and Prosper /alt3reg0 -- http://hem1.passagen.se/zman/main.html One of the two heavily cloaked and robed figures introduced himself as Neil, an elven magician. The other, who was left till last, finally stood and said, "I an Number Five." With the cloak pulled back, a mask could be seen over his face. Odd, thought Platos, I was under the impression that there were no other Greeks in the party, yet here is one of my superiors coming with us. Further thought along this line was forestalled by one of the masked men at the side coming forward. "Hey," he cried, "I'm Number Five !" The robed man turned instantly towards him and without losing his confidence said, "Well then, I must be Number 5.1 !" -- extract from "Prophecy" by Guy Simms. ------------------------------------------------------------------------ GET A NEXTCARD VISA, in 30 seconds! Get rates as low as 0.0% Intro APR and no hidden fees. Apply NOW! http://click.egroups.com/1/969/0/_/451227/_/951945546/ ------------------------------------------------------------------------ -> Spread the URL: http://www.onelist.com/subscribe/amiga-c <- @ENDNODE @NODE "Mail_11" " Re: 64 Bit Info() " From: Duane McDonnell Date: 03-01-00 22:28:53 Subject: @{b}Re: 64 Bit Info()@{ub} From: Duane McDonnell > REG ULONG bpb, blocks; >#define ONEGIG 1000000000L > blocks = (info->id_NumBlocks) - (info->id_NumBlocksUsed); > Bytesfree = 0L; > GigaBytesFree = 0L; > for(bpb = info->id_BytesPerBlock; bpb; bpb --){ > BytesFree += blocks; > if( BytesFree >= ONEGIG ){ > GigaBytesFree ++; > BytesFree -= ONEGIG; > } > } >This works, but requires 512, 1024 (or whatever blocksize) >number of itterations to execute... >I really need another perspective please..... Maybe you'll find the attachment useful. It's a 32x32=>64 multiplier function I wrote which works on all 68k CPUs and AmigaOS versions. It should solve the problem nicely. ------------------------------------------------------------------------ Registering a domain name is quick and easy. http://click.egroups.com/1/1611/0/_/451227/_/951945653/ ------------------------------------------------------------------------ -> Spread the URL: http://www.onelist.com/subscribe/amiga-c <- @ENDNODE @NODE "Mail_12" " Re: dos.library Close() won't work! HELP! " From: Christian Hattemer Date: 03-02-00 00:45:09 Subject: @{b}Re: dos.library Close() won't work! HELP!@{ub} From: Christian Hattemer On 01-Mär-00 Tim Hanson wrote: > However if you want to be realy safe, write a delete replacement that > creates a file with the same name of exactly the same size then delete > that. I think that should wipe the original data completely. I don't need such a tool because I'm the only person who uses the system, but the purge utility posted earlier here could be useful. It seems that pfs doesn't like very large files (several MBs) in the .deldir and hangs in a loop when listing it. I used to copy a large drawer an delete it again to flush the large files, but the purge utility is much more comfortable. Bye, Chris -- __ /// __ /// Amiga 4000T 040/40 Team *AMIGA* \\\/// \XX/ http://www.riednet.wh.tu-darmstadt.de/~chris/ Auch wenn auf dem Laptop DOS läuft wird das von den Verkehrsbetrieben noch nicht als Behinderung anerkannt, so daß er einer älteren Dame besser den Platz freimachen sollte. ------------------------------------------------------------------------ GET A NEXTCARD VISA, in 30 seconds! Get rates as low as 0.0% Intro APR and no hidden fees. Apply NOW! http://click.egroups.com/1/969/0/_/451227/_/951954673/ ------------------------------------------------------------------------ -> Spread the URL: http://www.onelist.com/subscribe/amiga-c <- @ENDNODE @NODE "Mail_13" " Re: 64 Bit Info() " From: Duane McDonnell Date: 03-02-00 06:32:54 Subject: @{b}Re: 64 Bit Info()@{ub} From: Duane McDonnell >Now here's the clencher, it has to work on version 37 >so you can forget the UMult64() in utility library.... I just found my C version of muls64, you might find it more useful than the asm one. ------------------------------------------------------------------------ DON'T HATE YOUR RATE! Get a NextCard Visa, in 30 seconds! Get rates as low as 0.0% Intro or 9.9% Fixed APR and no hidden fees. Apply NOW! http://click.egroups.com/1/2120/0/_/451227/_/951974734/ ------------------------------------------------------------------------ -> Spread the URL: http://www.onelist.com/subscribe/amiga-c <- @ENDNODE @NODE "Mail_14" " Re: 64 Bit Info() " From: Colin Wenzel Date: 03-02-00 09:02:29 Subject: @{b}Re: 64 Bit Info()@{ub} From: Colin Wenzel On 02-Mar-00, Duane McDonnell wrote: > From: Duane McDonnell -snip- > Maybe you'll find the attachment useful. It's a 32x32=>64 multiplier > function I wrote which works on all 68k CPUs and AmigaOS versions. > It should solve the problem nicely. > Thanks, i'll give it a whirl.... ------------------------------------- Colin Wenzel. Australia. EMAIL: mailto:colstv@hotkey.net.au URL: http://www.hotkey.net.au/~colstv/ ICQ: 17608330 AMIGA: 4000T, 060/50, 150Mb RAM, O.S 3.5 --------------------------------------- ------------------------------------------------------------------------ MAXIMIZE YOUR CARD, MINIMIZE YOUR RATE! Get a NextCard Visa, in 30 seconds! Get rates as low as 0.0% Intro or 9.9% Fixed APR and no hidden fees. Apply NOW! http://click.egroups.com/1/2122/0/_/451227/_/951984350/ ------------------------------------------------------------------------ -> Spread the URL: http://www.onelist.com/subscribe/amiga-c <- @ENDNODE @NODE "Mail_15" " Re: Re: dos.library Close() won't work! HELP! " From: tim@ramjam.u-net.com Date: 03-02-00 12:18:18 Subject: @{b}Re: Re: dos.library Close() won't work! HELP!@{ub} From: tim@ramjam.u-net.com Hi Christian > I don't need such a tool because I'm the only person who uses the system, > but the purge utility posted earlier here could be useful. > It seems that pfs doesn't like very large files (several MBs) in the .deldir > and hangs in a loop when listing it. Which version of PFS? I've received no complaints from my customers about this problem, nor have I encountered it myself. Tim -- ========================================================================== Ramjam Consultants Ltd, 28 Queensway, Caversham, Reading, RG4 6SQ, England e-mail: support@ramjam.u-net.com http://www.ramjam.u-net.com/ telephone: +44 (0)118 946 5940 fax: +44 (0)118 946 5941 ------------------------------------------------------------------------ GET A NEXTCARD VISA, in 30 seconds! Get rates as low as 0.0% Intro or 9.9% Fixed APR and no hidden fees. Apply NOW! http://click.egroups.com/1/937/0/_/451227/_/951995902/ ------------------------------------------------------------------------ -> Spread the URL: http://www.onelist.com/subscribe/amiga-c <- @ENDNODE @NODE "Mail_16" " Re: Re: dos.library Close() won't work! HELP! " From: "Francis G. Loch" Date: 03-02-00 15:13:57 Subject: @{b}Re: Re: dos.library Close() won't work! HELP!@{ub} From: "=?iso-8859-1?q?Francis=20G.=20Loch?=" Hi there, --- tim@ramjam.u-net.com wrote: > Hi Christian > > > I don't need such a tool because I'm the only > person who uses the system, > > but the purge utility posted earlier here could be > useful. > > It seems that pfs doesn't like very large files > (several MBs) in the .deldir > > and hangs in a loop when listing it. > > Which version of PFS? I've received no complaints > from my customers > about this problem, nor have I encountered it > myself. Just to say, I had a problem with PFS3 when deleting a lot of files, some of them several megabytes in size. I'm not sure what happened, but PFS caused the hard-drive to go into a read-only state and gave me a lot of problems afterwards. The only way I could resolve this problem was to reformat the hard drive. I haven't had any problems since then. I don't know if this will make any difference, but it was a 4 gig hard-drive with a Power Flyer (set to TD64) running on AmigaOS 3.5. Kind regards, Francis. ===== e-mail:- fgloch@yahoo.com web-site:- http://www.geocities.com/fgloch __________________________________________________ Do You Yahoo!? Talk to your friends online with Yahoo! Messenger. http://im.yahoo.com ------------------------------------------------------------------------ MAXIMIZE YOUR CARD, MINIMIZE YOUR RATE! Get a NextCard Visa, in 30 seconds! Get rates as low as 0.0% Intro or 9.9% Fixed APR and no hidden fees. Apply NOW! http://click.egroups.com/1/2122/0/_/451227/_/952006438/ ------------------------------------------------------------------------ -> Spread the URL: http://www.onelist.com/subscribe/amiga-c <- @ENDNODE @NODE "Mail_17" " Re: Internet communication " From: Roger Light Date: 03-02-00 15:39:22 Subject: @{b}Re: Internet communication@{ub} From: Roger Light Hi, Charles Barr wrote: > I want to add a feature to my program, which when it is started > contacts the site and finds out if there is a new version of itself. > > Anyone any ideas on how i could do this? Have a file on your web site that contains the version number of your program. eg. http://www.amigaforever.free-online.co.uk/myprogramversion where myprogramversion contains just the version number. Should be easy enough to parse that ;) Alternatively, you could make a file with a list of your programs in that each have a version number associated with them - this would save excess clutter if you plan to release lots of programs that have this feature. Regards, Roger ------------------------------------------------------------------------ GET A NEXTCARD VISA, in 30 seconds! Get rates as low as 0.0% Intro or 9.9% Fixed APR and no hidden fees. Apply NOW! http://click.egroups.com/1/937/0/_/451227/_/952007969/ ------------------------------------------------------------------------ -> Spread the URL: http://www.onelist.com/subscribe/amiga-c <- @ENDNODE @NODE "Mail_18" " Re: GCC Debug Help Needed " From: Gavin Kinsey Date: 03-02-00 18:00:40 Subject: @{b}Re: GCC Debug Help Needed@{ub} From: Gavin Kinsey I received a message from Matthew Fletcher, a reply follows, > > If the problem is the lack of amigaos hunk support can anyone mail me > a specs file setup for a diffrent linker/assembler that they know > works. I don't know whether this is the problem but to get amigaoshunk support you need to recompile binutils. configure with a target of m68k-amigaoshunk, then after installing that make a link "ln -s /GG/lib/gcc-lib/m68k-amigaos /GG/lib/gcc-lib/m68k-amigaoshunk" so that you can compile for it with "gcc -b m68k-amigaoshunk". I think this is right but it is a while ago since I set this up. HTH. -- I like pigs. Dogs look up to us. Cats look down on us. Pigs treat us as equals. - Winston Churchill Gavin Kinsey ------------------------------------------------------------------------ GET A NEXTCARD VISA, in 30 seconds! Get rates as low as 2.9% Intro or 9.9% Fixed APR and no hidden fees. Apply NOW! http://click.egroups.com/1/936/0/_/451227/_/952027200/ ------------------------------------------------------------------------ -> Spread the URL: http://www.onelist.com/subscribe/amiga-c <- @ENDNODE @NODE "Mail_19" " Re: StormC asm " From: Allan Odgaard Date: 03-02-00 21:38:54 Subject: @{b}Re: StormC asm@{ub} From: Allan Odgaard On 01-Mar-00, Johan Rasten wrote: > Is there a better way to send parameters to asm functions from C/C++ than > using globals? Yes -- you can either declare the function to receive register arguments, extern "C" VOID Function (register __a0 STRPTR str, register __d0 ULONG len); The 'extern "C"' is just to avoid name mangling. And most people use this define: #define REG(x) register __ ## x So that they can just write REG(a0) etc. The other possibility is simply to read the arguments (in the asm function) from the stack, which is where the C compiler normally stores them. E.g. in the above (without the register keywords) your function would be called with 'str' as 4(sp) and 'len' as 8(sp) AFAIR. > I'm also interested in retrieving the results of the func. The value you put in register D0, when the function return, is what the C compiler will use as result. > Can I inline asm in my C++ progs? [...] No -- what is it that you are writing which not only need the functions in assembler, but also inlined? Must be quite speed-critical, since you're going through all these troubles... ;-) Regards Allan ------------------------------------------------------------------------ MAXIMIZE YOUR CARD, MINIMIZE YOUR RATE! Get a NextCard Visa, in 30 seconds! Get rates as low as 0.0% Intro or 9.9% Fixed APR and no hidden fees. Apply NOW! http://click.egroups.com/1/2122/0/_/451227/_/952032316/ ------------------------------------------------------------------------ -> Spread the URL: http://www.onelist.com/subscribe/amiga-c <- @ENDNODE @NODE "Mail_20" " Re: StormC asm " From: Johan Rasten Date: 03-02-00 22:38:16 Subject: @{b}Re: StormC asm@{ub} From: Johan Rasten Hello Allan On 02-Mar-00, Allan Odgaard wrote: > No -- what is it that you are writing which not only need the functions in > assembler, but also inlined? Must be quite speed-critical, since you're > going through all these troubles... ;-) Well, actually it's just a turn-based brutal civ/col/settlers/warzone wargame, but I draw to a buffer and then C2P it to a DB screen which is quite slow =( and second, I really hate the win* approach to coding. And of course, I'm just a total speed freak that can't live knowing there's a faster and smaller solution to my problem ;-P Thanks, man. RegardZ /J Rasten -- http://hem1.passagen.se/zman/main.html :-# User wears braces ------------------------------------------------------------------------ DON'T HATE YOUR RATE! Get a NextCard Visa, in 30 seconds! Get rates as low as 0.0% Intro or 9.9% Fixed APR and no hidden fees. Apply NOW! http://click.egroups.com/1/2120/0/_/451227/_/952035980/ ------------------------------------------------------------------------ -> Spread the URL: http://www.onelist.com/subscribe/amiga-c <- @ENDNODE @NODE "Mail_21" " Re: dos.library Close() won't work! HELP! " From: Christian Hattemer Date: 03-03-00 12:46:30 Subject: @{b}Re: dos.library Close() won't work! HELP!@{ub} From: Christian Hattemer On 02-Mär-00 tim@ramjam.u-net.com wrote: >> It seems that pfs doesn't like very large files (several MBs) in the >> .deldir and hangs in a loop when listing it. > Which version of PFS? I've received no complaints from my customers > about this problem, nor have I encountered it myself. I'm using pfs 18.5. It first happened when I started using pfs with version 18.3, I copied some backup files to the HD and deleted them later. When I listed the .deldir sometime later it crashed. With 18.5 the history lists a bugfix for that, but it's not fully fixed, it doesn't crash anymore but hangs in a loop when listing the .deldir. If you listen carefully you can hear the HDD making repetitive sounds. The limit where the problem occurs is above 13.874.884 bytes and below 394.695.356 bytes, these were my two testfiles. Bye, Chris -- __ /// __ /// Amiga 4000T 040/40 Team *AMIGA* \\\/// \XX/ http://www.riednet.wh.tu-darmstadt.de/~chris/ How many software engineers does it take to change a light bulb? One, but if he changes it, the whole building will probably fall down. ------------------------------------------------------------------------ GET A NEXTCARD VISA, in 30 seconds! Get rates as low as 2.9% Intro or 9.9% Fixed APR and no hidden fees. Apply NOW! http://click.egroups.com/1/936/0/_/451227/_/952085154/ ------------------------------------------------------------------------ -> Spread the URL: http://www.onelist.com/subscribe/amiga-c <- @ENDNODE @NODE "Mail_22" " Re: Internet communication " From: Lee Atkins Date: 03-03-00 20:47:07 Subject: @{b}Re: Internet communication@{ub} From: Lee Atkins > From: Charles Barr > > Hello, > I want to add a feature to my program, which when it is started > contacts the site and finds out if there is a new version of itself. > > Anyone any ideas on how i could do this? I am currently writing an embedded tftp 'thing'. So far it dials up, connects, gets an IP address and just sits there 'Pinging' every 30 seconds. This has taken a LOT of reading and it doesn't do anything yet! The quickest way to do this would be to find a ready written bit of code and use that. There is oodles of source out there. Lee. ------------------------------------------------------------------------ DON'T HATE YOUR RATE! Get a NextCard Visa, in 30 seconds! Get rates as low as 0.0% Intro or 9.9% Fixed APR and no hidden fees. Apply NOW! http://click.egroups.com/1/2120/0/_/451227/_/952113133/ ------------------------------------------------------------------------ -> Spread the URL: http://www.onelist.com/subscribe/amiga-c <- @ENDNODE @NODE "Mail_23" " Re: dos.library Close() won't work! HELP! " From: W1bBle Date: 03-04-00 01:07:04 Subject: @{b}Re: dos.library Close() won't work! HELP!@{ub} From: W1bBle Hello Francis On 02-Mar-00, you wrote: > Just to say, I had a problem with PFS3 when deleting a > lot of files, some of them several megabytes in size. > I'm not sure what happened, but PFS caused the > hard-drive to go into a read-only state and gave me a > lot of problems afterwards. The only way I could > resolve this problem was to reformat the hard drive. I > haven't had any problems since then. > > I don't know if this will make any difference, but it > was a 4 gig hard-drive with a Power Flyer (set to > TD64) running on AmigaOS 3.5. I bought a Power Flyer too. I thought it was going to be the stuff only dreams are made of. It was, up to a point. The rest of the way, it sucked. After I installed it, my hard drive developed errors, my CD changer only worked when the CD you wanted to access was in the first slot, and none of the four IDE devices could work in harmony together, whatever the configuration. So I bit the bullet and ditched it, despite spending a good £60 or more on it. After that, my main HD died (cause of death unknown), and once I had it replaced, I haven't looked back since. IMHO, if you have a weird HD problem, rip out the Flyer first and then see how your life changes. At first I suspected PFS, but how wrong I was.... Apologies if this message is loaded with heartfelt bitterness! :) Regards Giles PS. More apologies for being way off topic -- <------------------------------------------------> | Giles Burdett __ ___ | | IRC nick = W1bBle /// (. .) | | /// ( v ) | | A1200T 200MHz PPC __ /// // \\ | | OS 3.1 / Linux-m68k \\\/// .X # X. | | \XX/ XX___XX | <------------------------------------------------> ------------------------------------------------------------------------ PERFORM CPR ON YOUR APR! Get a NextCard Visa, in 30 seconds! Get rates as low as 0.0% Intro or 9.9% Fixed APR and no hidden fees. Apply NOW! http://click.egroups.com/1/2121/0/_/451227/_/952132176/ ------------------------------------------------------------------------ -> Spread the URL: http://www.onelist.com/subscribe/amiga-c <- @ENDNODE @NODE "Mail_24" " Re: Linking PPC asm objects to compiled warpup objects (before it drifted onto arrays) " From: Martyn Capewell Date: 03-04-00 04:03:03 Subject: @{b}Re: Linking PPC asm objects to compiled warpup objects (before it drifted onto arrays)@{ub} From: Martyn Capewell Hello Alastair On 29-Feb-00, you wrote: > From: "Alastair M. Robinson" > > Hi Gabriele, > >>> C code compiled for PPC is usually much more efficient than compiled >>> 68k code, and hand-optimised assembler won't give anything like the >>> speedup you're used to from 68k programming. Since this is the Amiga-C >>> list, I recommend you do as much as possible in C! > >> Is this due to pipelining in the PPC processor? Id est: the compiler >> knows how to use it at the best? Anyhow I always write first all > > Up to a point yes, but it's more to do with the PPC having 32 registers to > play with. When a compiler runs out of registers it starts parking data on > the stack, which of course takes time. Also, the PPC registers are all general purpose. With the 68k separating registers between address and data, it makes it more difficult for the compiler to generate code which takes account of this optimally. Often, the compiler would run out of one type of register, and although the alternative type of registers were available, it would have to shift things around to use them. On the PPC a compiler can put data and pointers anywhere it wants. Regards -- - Martyn Capewell ... but you can't expect to wield supreme executive power just because some watery tart threw a sword at you! ------------------------------------------------------------------------ MAXIMIZE YOUR CARD, MINIMIZE YOUR RATE! Get a NextCard Visa, in 30 seconds! Get rates as low as 0.0% Intro or 9.9% Fixed APR and no hidden fees. Apply NOW! http://click.egroups.com/1/2122/0/_/451227/_/952139122/ ------------------------------------------------------------------------ -> Spread the URL: http://www.onelist.com/subscribe/amiga-c <- @ENDNODE @NODE "Mail_25" " Re: Re: Smaller executable sizes " From: Duane McDonnell Date: 03-05-00 10:35:56 Subject: @{b}Re: Re: Smaller executable sizes@{ub} From: Duane McDonnell >BTW, just in case you are interested, these are the C-functions I use. I >have NO idea what all the weird codes in SPrintF do, but it works :-> \x16\xc0\x4e\x75 are the asm opcodes for: move.b d0,(a3)+ 16c0 rts 4e75 which is the standard char stuffing hook for RawDoFmt(). >void SPrintF( STRPTR buffer, STRPTR fmt, ... ) >{ > APTR args; > args = (APTR)((&fmt) + 1); > RawDoFmt( fmt, args, (void (*))"\x16\xc0\x4e\x75", buffer ); >} If you want variable args, you should use stdarg.h instead. ------------------------------------------------------------------------ MAXIMIZE YOUR CARD, MINIMIZE YOUR RATE! Get a NextCard Visa, in 30 seconds! Get rates as low as 0.0% Intro or 9.9% Fixed APR and no hidden fees. Apply NOW! http://click.egroups.com/1/2122/0/_/451227/_/952322255/ ------------------------------------------------------------------------ -> Spread the URL: http://www.onelist.com/subscribe/amiga-c <- @ENDNODE @NODE "Mail_26" " STL... " From: Alexander Niven-Jenkins Date: 03-05-00 15:30:24 Subject: @{b}STL...@{ub} From: Alexander Niven-Jenkins Hiya, Has anyone got something like sgi's STL to work with Storm/HiSOFT any Amiga C++ compiler (not things like ecgs). Or if not, does any body have the source to the 'string' class that works. The one supplied with the STL is a bit more than I need. I started to write my own, but if somebody already has there's no pont in me doing it :-) Kind regards... Alex -- Cremlin Software - http://www.cremlinsoftware.org Founder, lead programmer - ghandi@cremlinsoftware.org Production, programmer - anj@cremlinsoftware.org ------------------------------------------------------------------------ DON'T HATE YOUR RATE! Get a NextCard Visa, in 30 seconds! Get rates as low as 0.0% Intro or 9.9% Fixed APR and no hidden fees. Apply NOW! http://click.egroups.com/1/2120/0/_/451227/_/952266636/ ------------------------------------------------------------------------ -> Spread the URL: http://www.onelist.com/subscribe/amiga-c <- @ENDNODE @NODE "Mail_27" " Smaller executable sizes " From: Niels de Koning Date: 03-05-00 15:32:42 Subject: @{b}Smaller executable sizes@{ub} From: Niels de Koning Hi Amiga-C people! I have written a small commandline tool and I am about to upload it to Aminet but I am not yet satisfied with the executable I created. I think the executable I created is too large. Call me crazy in this day and age where harddisks have long broken the multi-gigabyte barrier, but I /love/ commandline tools that are just a couple of kb in size :-) My question to all of you is: Do you have any tips that can make me save a couple of bytes in the final executable? The only thing I already did in my program is to use as many ROM calls as I can. Maybe a certain combination of compiler settings? (I use SAS) Maybe some cute programming tricks? Maybe another compiler generates smaller code? I can remember a discussion about optimization and VBCC late last year. Is VBCC good at generating small executables? Thanks! // Niels de Koning ____________________________________________________________________ Co-editor on 3DAddict's GFXZone http://www.gfxzone.org ------------------------------------------------------------------------ PERFORM CPR ON YOUR APR! Get a NextCard Visa, in 30 seconds! Get rates as low as 0.0% Intro or 9.9% Fixed APR and no hidden fees. Apply NOW! http://click.egroups.com/1/2121/0/_/451227/_/952267166/ ------------------------------------------------------------------------ -> Spread the URL: http://www.onelist.com/subscribe/amiga-c <- @ENDNODE @NODE "Mail_28" " Re: Smaller executable sizes " From: Colin Wenzel Date: 03-05-00 16:13:45 Subject: @{b}Re: Smaller executable sizes@{ub} From: Colin Wenzel On 06-Mar-00, Niels de Koning wrote: > From: Niels de Koning > > Hi Amiga-C people! > > I have written a small commandline tool and I am about to upload it > to Aminet but I am not yet satisfied with the executable I created. I > think the executable I created is too large. Call me crazy in this > day and age where harddisks have long broken the multi-gigabyte > barrier, but I /love/ commandline tools that are just a couple of kb > in size :-) > > My question to all of you is: Do you have any tips that can make me > save a couple of bytes in the final executable? > > The only thing I already did in my program is to use as many ROM > calls as I can. > > Maybe a certain combination of compiler settings? (I use SAS) > > Maybe some cute programming tricks? > > Maybe another compiler generates smaller code? I can remember a > discussion about optimization and VBCC late last year. Is VBCC good > at generating small executables? > Thanks! > // Niels de Koning The biggest killer is the format() functions, they are HUGE.... So, avoid using printf(), sprintf(), or any xxxxintf() functions if you can. To print text only without formatting, use puts(). I frequently see code with no formatting, but just text outputting using printf() like: printf(" Hello World\n"); You will probably save up to 4K-10K by avoiding format().... Of course, if you dont want to port your code, you could always use DOS calls directly, like: PutStr( text ); And, for C' formatting use exec's RawDoFormat() Here is an Amiga replacement for printf & sprintf...... (They are named: printf2() & sprintf2() to prevent collisions with c.lib versions ) Attached........ The next biggest killer is static buffers like: *char buffer[ 3000 ] = 0;* This will put it in the data segment as 3000 bytes of space & adds that to the program size... Use: *char buffer[ 3000 ];* This will put it in the dx section & will NOT add to the program size, but will be allocated at load time. In the startup code of your program, all dx data is automatically cleared anyway, so it is still all zeroed out for you too, so there is no need to initialize it yourself. ------------------------------------- Colin Wenzel. Australia. EMAIL: mailto:colstv@hotkey.net.au URL: http://www.hotkey.net.au/~colstv/ ICQ: 17608330 AMIGA: 4000T, 060/50, 150Mb RAM, O.S 3.5 --------------------------------------- ------------------------------------------------------------------------ GET A NEXTCARD VISA, in 30 seconds! Get rates as low as 0.0% Intro or 9.9% Fixed APR and no hidden fees. Apply NOW! http://click.egroups.com/1/937/0/_/451227/_/952269303/ ------------------------------------------------------------------------ -> Spread the URL: http://www.onelist.com/subscribe/amiga-c <- @ENDNODE @NODE "Mail_29" " Re: Smaller executable sizes " From: Niels de Koning Date: 03-05-00 17:38:43 Subject: @{b}Re: Smaller executable sizes@{ub} From: Niels de Koning Hello Colin, > The biggest killer is the format() functions, they are HUGE.... > > So, avoid using printf(), sprintf(), or any xxxxintf() functions if > you can. First of all, thanks for your tips. I really appreciate it. I am currently using functions in my program that use RawDoFmt() and VPrintf(). I wanted to test how the code you send would affect the size of my executable but I couldn't get it to work. ( I blame Micro$oft for that, they make compiling too easy :-) ) I assume that I need to convert the sources you send to object files, so that I can link my program with those object files. Unfortunately I don't know how to do this. I tried compiling them the same way I normally compile ".c" sources, apparently "#asm" is not a recognized command..... BTW, just in case you are interested, these are the C-functions I use. I have NO idea what all the weird codes in SPrintF do, but it works :-> I borrowed this code from the examples in some developer package..... void PrintF(char *FormatString,...) { (void)VPrintf(FormatString,(LONG *)&FormatString+1L); (void)Flush(Output()); } void SPrintF( STRPTR buffer, STRPTR fmt, ... ) { APTR args; args = (APTR)((&fmt) + 1); RawDoFmt( fmt, args, (void (*))"\x16\xc0\x4e\x75", buffer ); } > The next biggest killer is static buffers like: > > *char buffer[ 3000 ] = 0;* > > This will put it in the data segment as 3000 bytes of space > & adds that to the program size... > > Use: > > *char buffer[ 3000 ];* > > This will put it in the dx section & will NOT add to the > program size, but will be allocated at load time. I did not know this! I am not using it in this particular program but I do use it in a couple of other programs. Regards, // Niels de Koning ____________________________________________________________________ Co-editor on 3DAddict's GFXZone #http://www.gfxzone.org# ------------------------------------------------------------------------ PERFORM CPR ON YOUR APR! Get a NextCard Visa, in 30 seconds! Get rates as low as 0.0% Intro or 9.9% Fixed APR and no hidden fees. Apply NOW! http://click.egroups.com/1/2121/0/_/451227/_/952274701/ ------------------------------------------------------------------------ -> Spread the URL: http://www.onelist.com/subscribe/amiga-c <- @ENDNODE @NODE "Mail_30" " Re: Smaller executable sizes " From: Christian Hattemer Date: 03-05-00 17:54:24 Subject: @{b}Re: Smaller executable sizes@{ub} From: Christian Hattemer On 05-Mär-00 Niels de Koning wrote: > My question to all of you is: Do you have any tips that can make me save > a couple of bytes in the final executable? Here are two lists with tips: http://www.abarnett.demon.co.uk/tutorial.html http://www.ontek.com/mikey/optimization.html The book 'Code Complete' by Steve McConnel has two chapters dealing with optimization in theory and practice. It's a very recommandable book anyway. At http://www.cyberdynesystems.de/ you can get stormamiga.lib, a set of linker libraries written in asm. They are very small and fast. However they were written for use with StormC, I don't know if the format is compatible to SAS. > Maybe another compiler generates smaller code? I can remember a AFAIK does SAS makes quite good code, gcc is better but also needs more time for full optimization. Finally keep in mind that it's not worth to make your code unreadable by applying all mentioned tips unless you really need minimum size and maximum speed. Bye, Chris -- __ /// __ /// Amiga 4000T 040/40 Team *AMIGA* \\\/// \XX/ http://www.riednet.wh.tu-darmstadt.de/~chris/ Die schwache Prozessorleistung des Amiga, gleicht sein Benutzer mit dem Verstand aus! ------------------------------------------------------------------------ PERFORM CPR ON YOUR APR! Get a NextCard Visa, in 30 seconds! Get rates as low as 0.0% Intro or 9.9% Fixed APR and no hidden fees. Apply NOW! http://click.egroups.com/1/2121/0/_/451227/_/952275394/ ------------------------------------------------------------------------ -> Spread the URL: http://www.onelist.com/subscribe/amiga-c <- @ENDNODE @NODE "Mail_31" " Re: Smaller executable sizes " From: Allan Odgaard Date: 03-05-00 18:11:24 Subject: @{b}Re: Smaller executable sizes@{ub} From: Allan Odgaard On 05-Mar-00, Colin Wenzel wrote: >> [...] >> My question to all of you is: Do you have any tips that can make me >> save a couple of bytes in the final executable? It depends a lot on the program. You should go through the source and remove the redundancy that the compiler can't spot, for example by changing several similar lines to a loop, perhaps using a table for the 'differences'... >> Maybe a certain combination of compiler settings? (I use SAS) I'm using StormC with the StormAmiga.lib, this have special versions of printf/scanf which have no floating point support, and this saves several kilobytes. >> Maybe another compiler generates smaller code? I can remember a >> discussion about optimization and VBCC late last year. Is VBCC good >> at generating small executables? VBCC is good at generating fast code, and faster code is often larger due to aligning, unrolling, inlining etc. -- but I think the quality of the generated code for VBCC is better than the competitirs (e.g. being the only one with common subexpression elimination AFAIK) so it may still be smaller. And you can disable the optimization options which have a tendency of enlarging the code. Based on my experience with StormC then I'd say that a better compiler would often be able to generate executables at half the size, but SAS/C is probably a tad better than StormC when it comes to efficient code. I have only tested VBCC with relatively small programs, so I cannot say if my estimation is correct. I would be glad if someone would try to compile a big program with the various compilers and report the executeable size -- unfortunately all my own big sources are using C++ and often bound to StormC, that is, I've never been able to compile them with other compilers, maybe due to lack of includes etc. > The next biggest killer is static buffers like: > *char buffer[ 3000 ] = 0;* I think it should be 'char buffer [3000] = { }'. But it's true that it ends up in the data segment rather than the bss segment -- I did not know that and am quite amazed that the compiler is that stupid... OTOH you shouldn't use globals, nor statically allocated arrays ;-) > In the startup code of your program, all dx data is > automatically cleared anyway, so it is still all zeroed I don't think this is a valid assumption. Regards Allan -- URL: http://www.diku.dk/students/duff/ MS-DOS Multitasking: Crashing both hard-drives at once... ------------------------------------------------------------------------ GET A NEXTCARD VISA, in 30 seconds! Get rates as low as 0.0% Intro APR and no hidden fees. Apply NOW! http://click.egroups.com/1/975/0/_/451227/_/952280069/ ------------------------------------------------------------------------ -> Spread the URL: http://www.onelist.com/subscribe/amiga-c <- @ENDNODE @NODE "Mail_32" " Re: Smaller executable sizes " From: Ilkka Lehtoranta Date: 03-05-00 19:03:31 Subject: @{b}Re: Smaller executable sizes@{ub} From: Ilkka Lehtoranta On 05-Maa-00, Niels de Koning wrote: NK> My question to all of you is: Do you have any tips that can make me save NK> a couple of bytes in the final executable? Get HunkShort from Aminet. It can shrink executables but doesn't alter the code itself. ------------------------------------------------------------------------ GET A NEXTCARD VISA, in 30 seconds! Get rates as low as 0.0% Intro or 9.9% Fixed APR and no hidden fees. Apply NOW! http://click.egroups.com/1/937/0/_/451227/_/952279728/ ------------------------------------------------------------------------ -> Spread the URL: http://www.onelist.com/subscribe/amiga-c <- @ENDNODE @NODE "Mail_33" " 508,338 Search Engines and Directories " From: WORLD SUBMISSION CENTER Date: 03-05-00 20:51:16 Subject: @{b}508,338 Search Engines and Directories@{ub} From: "WORLD SUBMISSION CENTER" WORLD SUBMISSION CENTER Blast your site to 508,338+ Search Engines, Directories,Classifieds, Awards Sites, FFA's Free Every Weeks !! http://submit10000.com Submit to 1 FFA from AMEFFA Get a permanent exposure for 3 months Get your ad viewed 300+ times a day GUARANTEE !! http://ameffa.com Become A Reseller For Us And Earn 50% Commission ( 12.50$ ) From Sales Generated >From Your Site, No Investing, Apply Now It's Free !! http://submit10000.com Start Making 400.00$ The First Month Just By Surfing The Net.Double it every 3 Months !! http://submit10000.com ------------------------------------------------------------------------ GET A NEXTCARD VISA, in 30 seconds! Get rates as low as 0.0% Intro or 9.9% Fixed APR and no hidden fees. Apply NOW! http://click.egroups.com/1/937/0/_/451227/_/952275419/ ------------------------------------------------------------------------ -> Spread the URL: http://www.onelist.com/subscribe/amiga-c <- @ENDNODE @NODE "Mail_34" " Re: Smaller executable size (again) " From: GURER OZEN Date: 03-05-00 21:07:16 Subject: @{b}Re: Smaller executable size (again)@{ub} From: "GURER OZEN" Hello, NdK-> know what I mean). I also noticed that my own program has a lot of NdK-> trash at the end of it. For example, there is text in there that says NdK-> "con:10/10/320/80". Does anybody know what this is for and how I can This is the console window for your program if it started from workbench. It comes from c.o startup module. If you don't need extra startup features of sas/c you can change your startup code with STARTUP=blala.c command line option. Available startup modules are in sasc:lib/ directory. You can also use NOSTARTUP switch for completely disabling startup code. But then you can't use stdio, stdout, argc&argv, etc without initializing. And you need to open/close system libraries yourself. Other usefull sas/c options for you: OPTIMIZE OPTIMIZESIZE (if you dont care about speed) PARAMETERS=REGISTERS NOSTACKCHECK -- Gurer Ozen (gozen@isbank.net.tr, palpa@linuxfan.com) Si non confectus, non reficiat! To seek freedom is the only driving force I know. Freedom to lift off; to be like the flame of a candle, which, in spite of being up against the light of a billion stars, remains intact, because it never pretended to be more than what it is: a mere candle. -- Nagual Don Juan ------------------------------------------------------------------------ PERFORM CPR ON YOUR APR! Get a NextCard Visa, in 30 seconds! Get rates as low as 0.0% Intro or 9.9% Fixed APR and no hidden fees. Apply NOW! http://click.egroups.com/1/2121/0/_/451227/_/952290474/ ------------------------------------------------------------------------ -> Spread the URL: http://www.onelist.com/subscribe/amiga-c <- @ENDNODE @NODE "Mail_35" " Smaller executable size (again) " From: Niels de Koning Date: 03-05-00 21:18:22 Subject: @{b}Smaller executable size (again)@{ub} From: Niels de Koning Hi again, Amiga-C people! I realize my last question about smaller executables sizes has been a bit vague because I didn't really know where to look for those precious extra bytes. So far the answers I got have shaved off a couple of 100 bytes from my executable!! The URLs Christian Hattemer send on the topic of optimizations even made my routines a couple of percent faster :-) This mail isn't going to be any less vague than the previous one, but I thought I could at least try to be a bit more specific. Now I have stumbled on something that /seems/ to be what I want. I have been looking at my own executable and an executable from the C: directory. I took C:Avail as comparison. This program is a whopping 736 bytes large but it takes commandline parameters and generates output! When I look at the hex-representation of my executable and C:Avail I notice that the code of C:Avail looks much more 'compact' (I hope you know what I mean). I also noticed that my own program has a lot of trash at the end of it. For example, there is text in there that says "con:10/10/320/80". Does anybody know what this is for and how I can get rid of it? Apparently it is not needed, otherwise C:Avail would have it as well, right? When I create a very small C-program that does completely nothing (In the main I declare a variable which I initialize to 10 and return) The resulting executable is over 3 times as large as C:Avail. Yet this program does a lot more :-) So MY executables seem to have a certain /minimum/ size, yet C:Avail is a lot smaller than this. Does anybody know how C:Avail is compiled? I am not looking for a detailed explanation ofcourse but I am quite curious about this neat trick. >From my point of view it looks as if there is a way to use a smaller 'header' for an executable or something similar ?!? These are all issues in an unknown area of programming for me. I hope you can shed some light on it. Thanks! // Niels de Koning ____________________________________________________________________ Co-editor on 3DAddict's GFXZone http://www.gfxzone.org ------------------------------------------------------------------------ PERFORM CPR ON YOUR APR! Get a NextCard Visa, in 30 seconds! Get rates as low as 0.0% Intro or 9.9% Fixed APR and no hidden fees. Apply NOW! http://click.egroups.com/1/2121/0/_/451227/_/952287879/ ------------------------------------------------------------------------ -> Spread the URL: http://www.onelist.com/subscribe/amiga-c <- @ENDNODE @NODE "Mail_36" " Re: Smaller executable size (again) " From: Ilkka Lehtoranta Date: 03-05-00 21:31:52 Subject: @{b}Re: Smaller executable size (again)@{ub} From: Ilkka Lehtoranta On 05-Maa-00, Niels de Koning wrote: NK> When I look at the hex-representation of my executable and C:Avail I NK> notice that the code of C:Avail looks much more 'compact' (I hope you NK> know what I mean). I also noticed that my own program has a lot of NK> trash at the end of it. For example, there is text in there that says NK> "con:10/10/320/80". Does anybody know what this is for and how I can NK> get rid of it? Apparently it is not needed, otherwise C:Avail would NK> have it as well, right? I guess you are using a different startup module. IRC there are several options for startup modules depending on is your program CLI only. For example, Workbench startup messages need some code but isn't needed for CLI. And in WB you have no output window so startup modules contain some code to open it automatically if required (e.g. "library x not available"). Certainly not required if executed from CLI. IRC you can disable useless WB features somehow... But can't remember just now any more :I ------------------------------------------------------------------------ GET A NEXTCARD VISA, in 30 seconds! Get rates as low as 0.0% Intro or 9.9% Fixed APR and no hidden fees. Apply NOW! http://click.egroups.com/1/937/0/_/451227/_/952288635/ ------------------------------------------------------------------------ -> Spread the URL: http://www.onelist.com/subscribe/amiga-c <- @ENDNODE @NODE "Mail_37" " OT: Missing Digests - a plea " From: "Alastair M. Robinson" Date: 03-05-00 21:42:37 Subject: @{b}OT: Missing Digests - a plea@{ub} From: "Alastair M. Robinson" Hi everyone, I've now had no fewer than eight digests from this list go missing in the last fortnight, so I'm now remonstrating with my mailbox provider. I wondered if the problem might be due to some over-zealous server somewhere spotting the OneList sponsor ad and assuming the mails were spam? Anyway, could someone *please* send me an lzx'ed copy of some or all of Digests 289, 294, 295, 297, 304, 305, 306 and 307! I can then try sending copies to myself and see if anything guns them down. (BTW Allan, I couldn't find any way of collecting digests from OneList's site, so if no-one answers this plea, I'll just have to wait for the Aminet archive.) All the best, -- Alastair M. Robinson, email:blackfive@fakenhamweb.co.uk Amiga 4000/030, 18Meg RAM, 4.3Gig HD, Mustek ScanExpress 6000SP C, Assembler (68k, PIC16C84, ATMEL AVR), PostScript, ARexx, AMOSPro(!), AmigaDOS 9 out of 10 cats prefer hamsters. ------------------------------------------------------------------------ GET A NEXTCARD VISA, in 30 seconds! Get rates as low as 2.9% Intro or 9.9% Fixed APR and no hidden fees. Apply NOW! http://click.egroups.com/1/936/0/_/451227/_/952288933/ ------------------------------------------------------------------------ -> Spread the URL: http://www.onelist.com/subscribe/amiga-c <- @ENDNODE @NODE "Mail_38" " Re: Smaller exe size " From: Colin Wenzel Date: 03-06-00 02:52:08 Subject: @{b}Re: Smaller exe size@{ub} From: Colin Wenzel On 06-Mar-00, Allan Odgaard wrote: > From: Allan Odgaard > On 05-Mar-00, Colin Wenzel wrote: -snip- > I think it should be 'char buffer [3000] = { }'. But it's true that > it ends up in the data segment rather than the bss segment -- I did > not know that and am quite amazed that the compiler is that stupid... > OTOH you shouldn't use globals, nor statically allocated arrays ;-) > >> In the startup code of your program, all dx data is >> automatically cleared anyway, so it is still all zeroed > > I don't think this is a valid assumption. > I use Aztec C' (K&R) that's probably the syntax difference. The clearDX routine is in the link module: crt0.c ( the .begin routine.) Check crt0.c source, I believe it is mandatory . ------------------------------------- Colin Wenzel. Australia. EMAIL: mailto:colstv@hotkey.net.au URL: http://www.hotkey.net.au/~colstv/ ICQ: 17608330 AMIGA: 4000T, 060/50, 150Mb RAM, O.S 3.5 --------------------------------------- ------------------------------------------------------------------------ GET A NEXTCARD VISA, in 30 seconds! Get rates as low as 0.0% Intro or 9.9% Fixed APR and no hidden fees. Apply NOW! http://click.egroups.com/1/937/0/_/451227/_/952307607/ ------------------------------------------------------------------------ -> Spread the URL: http://www.onelist.com/subscribe/amiga-c <- @ENDNODE @NODE "Mail_39" " Re: Smaller exe size (again) " From: Colin Wenzel Date: 03-06-00 02:52:19 Subject: @{b}Re: Smaller exe size (again)@{ub} From: Colin Wenzel On 06-Mar-00, Ilkka Lehtoranta wrote: > I guess you are using a different startup module. IRC there are > several options for startup modules depending on is your program CLI > only. For example, Workbench startup messages need some code but > isn't needed for CLI. And in WB you have no output window so startup > modules contain some code to open it automatically if required (e.g. > "library x not available"). Certainly not required if executed from > CLI. IRC you can disable useless WB features somehow... But can't > remember just now any more :I To disable the WB startup icon parser & stuff you dont need for cli programs (you can still start from WB, but you get no args) declare: * -wb-parse() * // the '-' is an underscore (damn mailer commands) make a null function: * -wb-parse(void) {;}* Your compiler may differ, no warranties or responsibility for any threpson & blither nethapon wheras the frippet your diphon flange. All parties expressly forbid the ultim drether & respond to it, where upon the resdon & desperation of people reading this drivvel. ------------------------------------- Colin Wenzel. Australia. EMAIL: mailto:colstv@hotkey.net.au URL: http://www.hotkey.net.au/~colstv/ ICQ: 17608330 AMIGA: 4000T, 060/50, 150Mb RAM, O.S 3.5 --------------------------------------- ------------------------------------------------------------------------ GET A NEXTCARD VISA, in 30 seconds! Get rates as low as 0.0% Intro or 9.9% Fixed APR and no hidden fees. Apply NOW! http://click.egroups.com/1/937/0/_/451227/_/952307605/ ------------------------------------------------------------------------ -> Spread the URL: http://www.onelist.com/subscribe/amiga-c <- @ENDNODE @NODE "Mail_40" " PPC bugs with file access " From: Steven Dobbs Date: 03-06-00 09:21:40 Subject: @{b}PPC bugs with file access@{ub} From: Steven Dobbs Hello it seems that my system is prone to crashing when I use ansi-c file access stuff. It seems to happen at or soon after the close(fp) thing. I've been using both vbcc and stormc v3.0, both for warpos. Is this a common problem I wonder, or one particular to my system? is there a solution? There are of course, no problems when compiling for 040. Regards -- Steven Dobbs "AMIGA Farm" http://web.ukonline.co.uk/donkeedobbs ------------------------------------------------------------------------ DON'T HATE YOUR RATE! Get a NextCard Visa, in 30 seconds! Get rates as low as 0.0% Intro or 9.9% Fixed APR and no hidden fees. Apply NOW! http://click.egroups.com/1/2120/0/_/451227/_/952330975/ ------------------------------------------------------------------------ -> Spread the URL: http://www.onelist.com/subscribe/amiga-c <- @ENDNODE @NODE "Mail_41" " Re: Re: dos.library Close() won't work! HELP! " From: "Francis G. Loch" Date: 03-06-00 09:56:37 Subject: @{b}Re: Re: dos.library Close() won't work! HELP!@{ub} From: "=?iso-8859-1?q?Francis=20G.=20Loch?=" Hi W1bBle, --- W1bBle wrote: > Hello Francis > > On 02-Mar-00, you wrote: > > Just to say, I had a problem with PFS3 when > deleting a > > lot of files, some of them several megabytes in > size. > > I'm not sure what happened, but PFS caused the > > hard-drive to go into a read-only state and gave > me a > > lot of problems afterwards. The only way I could > > resolve this problem was to reformat the hard > drive. I > > haven't had any problems since then. > > > > I don't know if this will make any difference, but > it > > was a 4 gig hard-drive with a Power Flyer (set to > > TD64) running on AmigaOS 3.5. > > I bought a Power Flyer too. I thought it was going > to be the stuff only > dreams are made of. It was, up to a point. The rest > of the way, it sucked. > After I installed it, my hard drive developed > errors, my CD changer only > worked when the CD you wanted to access was in the > first slot, and none of > the four IDE devices could work in harmony together, > whatever the > configuration. So I bit the bullet and ditched it, > despite spending a good > £60 or more on it. After that, my main HD died > (cause of death unknown), > and once I had it replaced, I haven't looked back > since. > IMHO, if you have a weird HD problem, rip out the > Flyer first and then see > how your life changes. At first I suspected PFS, but > how wrong I was.... > > Apologies if this message is loaded with heartfelt > bitterness! :) Hmm, it sounds as though you didn't have a good time with the Flyer. I got mine when they were just released and I have never had any problems it. The only problem I ever had with it is trying to get it to fit nicely with my BVisionPPC card. Kind regards, Francis. ===== e-mail:- fgloch@yahoo.com web-site:- http://www.geocities.com/fgloch __________________________________________________ Do You Yahoo!? Talk to your friends online with Yahoo! Messenger. http://im.yahoo.com ------------------------------------------------------------------------ DON'T HATE YOUR RATE! Get a NextCard Visa, in 30 seconds! Get rates as low as 0.0% Intro or 9.9% Fixed APR and no hidden fees. Apply NOW! http://click.egroups.com/1/2120/0/_/451227/_/952332997/ ------------------------------------------------------------------------ -> Spread the URL: http://www.onelist.com/subscribe/amiga-c <- @ENDNODE @NODE "Mail_42" " Re: GCC Debug Help Needed " From: Matthew Fletcher Date: 03-06-00 11:28:36 Subject: @{b}Re: GCC Debug Help Needed@{ub} From: "Matthew Fletcher" --- In amiga-c@onelist.com, Gavin Kinsey wrote: > I don't know whether this is the problem but to get amigaoshunk > support you need to recompile binutils. configure with a target of > m68k-amigaoshunk, then after installing that make a link "ln -s > /GG/lib/gcc-lib/m68k-amigaos /GG/lib/gcc- lib/m68k-amigaoshunk" > so that you can compile for it with "gcc -b m68k-amigaoshunk". I was kinda hopeing it wouldent come to that, anyone know how to setup a specs file so i can use a diffrent assembler / linker. Or any debugers that support stabs for source debugging. -matthew ------------------------------------------------------------------------ DON'T HATE YOUR RATE! Get a NextCard Visa, in 30 seconds! Get rates as low as 0.0% Intro or 9.9% Fixed APR and no hidden fees. Apply NOW! http://click.egroups.com/1/2120/0/_/451227/_/952338519/ ------------------------------------------------------------------------ -> Spread the URL: http://www.onelist.com/subscribe/amiga-c <- @ENDNODE @NODE "Mail_43" " RE: Internet communication " From: roderick_maclean@agilent.com Date: 03-06-00 12:26:14 Subject: @{b}RE: Internet communication@{ub} From: roderick_maclean@agilent.com Charles Barr wrote: >> I want to add a feature to my program, which when it is started >> contacts the site and finds out if there is a new version of itself. >> >> Anyone any ideas on how i could do this? Lee Atkins wrote: > The quickest way to do this would be to find a ready written > bit of code and > use that. There is oodles of source out there. I'm not entirely sure if this will help but the code for ADT (Aminet Download Tool) does the following (IIRC): o- downloads the INDEX file o- compares it to something o- lets you download other stuff... You need to download a file, compare internal version to the version contained in the file. You can then decide whether to make another download (the latest version of the program) based upon that information. So maybe some of the ADT code will help? It's on Aminet ;) I hope this helps! Rod -- Rod MacLean | Roderick_MacLean@agilent.com Developer - Agilent Technologies | +44 (0131) 331-7304 ** My views do not necessarily reflect the views of Agilent! ** ------------------------------------------------------------------------ GET A NEXTCARD VISA, in 30 seconds! Get rates as low as 0.0% Intro or 9.9% Fixed APR and no hidden fees. Apply NOW! http://click.egroups.com/1/937/0/_/451227/_/952341998/ ------------------------------------------------------------------------ -> Spread the URL: http://www.onelist.com/subscribe/amiga-c <- @ENDNODE @NODE "Mail_44" " 508,338 Search Engines and Directories " From: WORLD SUBMISSION CENTER Date: 03-06-00 12:33:43 Subject: @{b}508,338 Search Engines and Directories@{ub} From: "WORLD SUBMISSION CENTER" WORLD SUBMISSION CENTER Blast your site to 508,338+ Search Engines, Directories,Classifieds, Awards Sites, FFA's Free Every Weeks !! http://submit10000.com Submit to 1 FFA from AMEFFA Get a permanent exposure for 3 months Get your ad viewed 300+ times a day GUARANTEE !! http://ameffa.com Become A Reseller For Us And Earn 50% Commission ( 12.50$ ) From Sales Generated >From Your Site, No Investing, Apply Now It's Free !! http://submit10000.com Start Making 400.00$ The First Month Just By Surfing The Net.Double it every 3 Months !! http://submit10000.com ------------------------------------------------------------------------ GET A NEXTCARD VISA, in 30 seconds! Get rates as low as 0.0% Intro or 9.9% Fixed APR and no hidden fees. Apply NOW! http://click.egroups.com/1/937/0/_/451227/_/952331976/ ------------------------------------------------------------------------ -> Spread the URL: http://www.onelist.com/subscribe/amiga-c <- @ENDNODE @NODE "Mail_45" " Re: Smaller executable sizes " From: Tim Hanson Date: 03-06-00 13:35:12 Subject: @{b}Re: Smaller executable sizes@{ub} From: Tim Hanson Hi Allan Allan Odgaard wrote this on 05-Mar-00 at 18:11:24 > VBCC is good at generating fast code, and faster code is often larger due to > aligning, unrolling, inlining etc. -- but I think the quality of the > generated code for VBCC is better than the competitirs (e.g. being the only > one with common subexpression elimination AFAIK) so it may still be smaller. > And you can disable the optimization options which have a tendency of > enlarging the code. I`ve found that Vbcc creates larger exes than Dice but a hell of a lot faster. Even without optimisation on. -- Tim Hanson, by the River,Canterbury, Kent. Amiga A1200T,`040/25,603e/160,32MB Fast,BVisionPPC, 8.4GB IDE HD,ZIP SCSI,RENO x2CD SCSI,Pace 56K Don`t waste your time by visiting http://www.sinister67.freeserve.co.uk ------------------------------------------------------------------------ DON'T HATE YOUR RATE! Get a NextCard Visa, in 30 seconds! Get rates as low as 0.0% Intro or 9.9% Fixed APR and no hidden fees. Apply NOW! http://click.egroups.com/1/2120/0/_/451227/_/952371741/ ------------------------------------------------------------------------ -> Spread the URL: http://www.onelist.com/subscribe/amiga-c <- @ENDNODE @NODE "Mail_46" " Re: Missing Digests - a plea " From: Tim Hanson Date: 03-06-00 13:40:10 Subject: @{b}Re: Missing Digests - a plea@{ub} From: Tim Hanson Hi Alastair Alastair M. Robinson wrote this on 05-Mar-00 at 21:42:37 > (BTW Allan, I couldn't find any way of collecting digests from OneList's > site, so if no-one answers this plea, I'll just have to wait for the Aminet > archive.) The aminet archive was sent to de.aminet.net on friday so it should be there soon. If not and your in a hurry I could send you it privately. -- Tim Hanson, by the River,Canterbury, Kent. Amiga A1200T,`040/25,603e/160,32MB Fast,BVisionPPC, 8.4GB IDE HD,ZIP SCSI,RENO x2CD SCSI,Pace 56K Don`t waste your time by visiting http://www.sinister67.freeserve.co.uk ------------------------------------------------------------------------ GET A NEXTCARD VISA, in 30 seconds! Get rates as low as 2.9% Intro or 9.9% Fixed APR and no hidden fees. Apply NOW! http://click.egroups.com/1/936/0/_/451227/_/952371739/ ------------------------------------------------------------------------ -> Spread the URL: http://www.onelist.com/subscribe/amiga-c <- @ENDNODE @NODE "Mail_47" " Re: Smaller exe size " From: Allan Odgaard Date: 03-06-00 14:06:15 Subject: @{b}Re: Smaller exe size@{ub} From: Allan Odgaard On 06-Mar-00, Colin Wenzel wrote: > The clearDX routine is in the link module: crt0.c > ( the .begin routine.) But it may be limited to the compiler you use. Earlier I also relied on everything new'ed to be cleared, because that's the case for StormC. And that resulted in some problems when I had to run some sources under unix -- my problem wasn't serious as I could simply overload the new operator. But had it been globals, then I'd have a problem, as I have no idea about startup code on unix etc. Speaking of unix -- I'm using LEDA for a project, and even small files take forever to link. So I was wondering if perhaps unix link libraries can be created with and without an index, and in the latter case they are slow (proportional to their size). Can anybody clarify this for me? Regards Allan -- URL: http://www.diku.dk/students/duff/ You mean you have Windows98 and still have available memory?!? ------------------------------------------------------------------------ GET A NEXTCARD VISA, in 30 seconds! Get rates as low as 0.0% Intro or 9.9% Fixed APR and no hidden fees. Apply NOW! http://click.egroups.com/1/937/0/_/451227/_/952349619/ ------------------------------------------------------------------------ -> Spread the URL: http://www.onelist.com/subscribe/amiga-c <- @ENDNODE @NODE "Mail_48" " Re: Missing Digests - a plea " From: Allan Odgaard Date: 03-06-00 14:14:53 Subject: @{b}Re: Missing Digests - a plea@{ub} From: Allan Odgaard On 05-Mar-00, Alastair M. Robinson wrote: > (BTW Allan, I couldn't find any way of collecting digests from OneList's All I know is that when I perform a search on Onelist, I'm presented with links to the digest that contain my keywords -- so you could e.g. search for 'feb'. Regards Allan -- URL: http://www.diku.dk/students/duff/ Buy Pentium Pro 400MHz! You reboot faster... ------------------------------------------------------------------------ GET A NEXTCARD VISA, in 30 seconds! Get rates as low as 2.9% Intro or 9.9% Fixed APR and no hidden fees. Apply NOW! http://click.egroups.com/1/936/0/_/451227/_/952349621/ ------------------------------------------------------------------------ -> Spread the URL: http://www.onelist.com/subscribe/amiga-c <- @ENDNODE @NODE "Mail_49" " RE: Smaller executable size (again) " From: roderick_maclean@agilent.com Date: 03-06-00 18:35:12 Subject: @{b}RE: Smaller executable size (again)@{ub} From: roderick_maclean@agilent.com > NdK-> know what I mean). I also noticed that my own program > has a lot of > NdK-> trash at the end of it. For example, there is text in > there that says > NdK-> "con:10/10/320/80". Does anybody know what this is for > and how I can > > This is the console window for your program if it started > from workbench. > It comes from c.o startup module. > > If you don't need extra startup features of sas/c you can change your > startup code with STARTUP=blala.c command line option. Available > startup modules are in sasc:lib/ directory. You can also use NOSTARTUP > switch for completely disabling startup code. But then you can't use > stdio, stdout, argc&argv, etc without initializing. And you > need to open/close > system libraries yourself. > > Other usefull sas/c options for you: > OPTIMIZE > OPTIMIZESIZE (if you dont care about speed) > PARAMETERS=REGISTERS > NOSTACKCHECK > There is an SAS/C example (concat, IIRC) which shows you how to create an executable without any startup code. It shows you how to open DOS.library, (exec as well, I think) and also use argc & argv, open stdio streams etc. I've experimented with opening these things myself from within my C code and it turns out to be a lot smaller than the SAS/C startup code ;) It's one of these pieces of code that once you've got it working, you never need to change :) What's more, there are no special pre-processor switches or pre-compiler directives so it *might* work under other Amiga C compilers. (But don't count on it). Rod -- Rod MacLean | Roderick_MacLean@agilent.com Developer - Agilent Technologies | +44 (0131) 331-7304 ** My views do not necessarily reflect the views of Agilent! ** ------------------------------------------------------------------------ GET A NEXTCARD VISA, in 30 seconds! Get rates as low as 0.0% Intro or 9.9% Fixed APR and no hidden fees. Apply NOW! http://click.egroups.com/1/937/0/_/451227/_/952364131/ ------------------------------------------------------------------------ -> Spread the URL: http://www.onelist.com/subscribe/amiga-c <- @ENDNODE @NODE "Mail_50" " RE: PPC bugs with file access " From: Jesper Svennevid Date: 03-06-00 21:58:44 Subject: @{b}RE: PPC bugs with file access@{ub} From: Jesper Svennevid > -----Original Message----- > From: Steven Dobbs [mailto:steven.dobbs@cableinet.co.uk] > Sent: Monday, March 06, 2000 9:22 AM > To: amiga-c@onelist.com > Subject: [amiga-c] PPC bugs with file access > > > From: Steven Dobbs > > Hello > it seems that my system is prone to crashing when I use > ansi-c file > access stuff. It seems to happen at or soon after the close(fp) thing. > > I've been using both vbcc and stormc v3.0, both for warpos. > Is this a common > problem I wonder, or one particular to my system? is there a > solution? > > There are of course, no problems when compiling for 040. I have no problems with ansi-c filehandling under warpos/powerup. I use egcs 1.1.2 (2.91.66) for this, and it works just great, there must be something else interfering. -- // Jesper Svennevid : jesper@fxrealm.com Software Developer at FXRealm Studios http://www.fxrealm.com/ ------------------------------------------------------------------------ GET A NEXTCARD VISA, in 30 seconds! Get rates as low as 0.0% Intro APR and no hidden fees. Apply NOW! http://click.egroups.com/1/975/0/_/451227/_/952376382/ ------------------------------------------------------------------------ -> Spread the URL: http://www.onelist.com/subscribe/amiga-c <- @ENDNODE @NODE "Mail_51" " Re: Re: Smaller exe size " From: Volker Barthelmann Date: 03-07-00 01:18:36 Subject: @{b}Re: Re: Smaller exe size@{ub} From: Volker Barthelmann On Mon, 6 Mar 2000, Allan Odgaard wrote: > On 06-Mar-00, Colin Wenzel wrote: > > > The clearDX routine is in the link module: crt0.c > > ( the .begin routine.) AmigaOS zeroes the bss-segment when loading an executable. Might be very old versions didn't do this, but there's no need to do this in the startup-code these days. Usually this only has to be done on systems running out of ROM (e.g. in some embedded systems), not for dynamically loaded code. > But it may be limited to the compiler you use. Not for static variables. C specifies that all parts of objects with static storage-duration (i.e. variables defined with "static" or with file-scope) which are not explicitly initialized, will get initialized to all 0 once before the program is started. There are some exceptions with unions if different members have different bit-patterns for 0, but this does not apply to the Amiga. > Earlier I also relied on everything new'ed to be cleared, because that's the > case for StormC. And that resulted in some problems when I had to run some > sources under unix -- my problem wasn't serious as I could simply overload the > new operator. This is something different. Automatic variables (i.e. variables defined within functions without "static") and memory returned by malloc are not initialized to zero (unless explicitly done by the application). Volker ------------------------------------------------------------------------ GET A NEXTCARD VISA, in 30 seconds! Get rates as low as 0.0% Intro or 9.9% Fixed APR and no hidden fees. Apply NOW! http://click.egroups.com/1/937/0/_/451227/_/952394631/ ------------------------------------------------------------------------ -> Spread the URL: http://www.onelist.com/subscribe/amiga-c <- @ENDNODE @NODE "Mail_52" " Re: PPC bugs with file access " From: Stefan Schulze Date: 03-07-00 01:23:14 Subject: @{b}Re: PPC bugs with file access@{ub} From: Stefan Schulze Hi Steven, > I've been using both vbcc and stormc v3.0, both for warpos. Is this a common > problem I wonder, or one particular to my system? is there a solution? I use StormC V3 and the mentioned instructions work fine here. Bye, Stefan. -- "The dream has led me - And i will follow into the glowing fire." (Lacrimosa) Dreamworlds Development - http://www.dreamworlds.de XFire I/II, YATAG, VoxelRace, Between The Lines ------------------------------------------------------------------------ MAXIMIZE YOUR CARD, MINIMIZE YOUR RATE! Get a NextCard Visa, in 30 seconds! Get rates as low as 0.0% Intro or 9.9% Fixed APR and no hidden fees. Apply NOW! http://click.egroups.com/1/2122/0/_/451227/_/952389044/ ------------------------------------------------------------------------ -> Spread the URL: http://www.onelist.com/subscribe/amiga-c <- @ENDNODE @NODE "Mail_53" " 508,338 Search Engines and Directories " From: "508.338 Search Engines and Directories" Date: 03-07-00 11:32:15 Subject: @{b}508,338 Search Engines and Directories@{ub} From: "508.338 Search Engines and Directories" WORLD SUBMISSION CENTER Blast your site to 508,338+ Search Engines, Directories,Classifieds, Awards Sites, FFA's Free Every Weeks !! http://submit10000.com Submit to 1 FFA from AMEFFA Get a permanent exposure for 3 months Get your ad viewed 300+ times a day GUARANTEE !! http://ameffa.com Become A Reseller For Us And Earn 50% Commission ( 12.50$ ) From Sales Generated >From Your Site, No Investing, Apply Now It's Free !! http://submit10000.com Start Making 400.00$ The First Month Just By Surfing The Net.Double it every 3 Months !! http://submit10000.com ------------------------------------------------------------------------ MAXIMIZE YOUR CARD, MINIMIZE YOUR RATE! Get a NextCard Visa, in 30 seconds! Get rates as low as 0.0% Intro or 9.9% Fixed APR and no hidden fees. Apply NOW! http://click.egroups.com/1/2122/0/_/451227/_/952414689/ ------------------------------------------------------------------------ -> Spread the URL: http://www.onelist.com/subscribe/amiga-c <- @ENDNODE @NODE "Mail_54" " RE: 508,338 Search Engines and Directories " From: roderick_maclean@agilent.com Date: 03-07-00 16:02:22 Subject: @{b}RE: 508,338 Search Engines and Directories@{ub} From: roderick_maclean@agilent.com > Date: Tue, 07 Mar 2000 02:32:15 -0800 > From: "508.338 Search Engines and Directories" > > Subject: 508,338 Search Engines and Directories Is there any way that we can prevent spammers from sending junk to this list? This is the second piece of spam in two days. The next thing they'll try is ripping our email addresses out of the posts and I don't really want that to happen - this is my work address!! I don't want to unsubscribe from the list either. Rod -- Rod MacLean | Roderick_MacLean@agilent.com Developer - Agilent Technologies | +44 (0131) 331-7304 ** My views do not necessarily reflect the views of Agilent! ** ------------------------------------------------------------------------ GET A NEXTCARD VISA, in 30 seconds! Get rates as low as 0.0% Intro APR and no hidden fees. Apply NOW! http://click.egroups.com/1/975/0/_/451227/_/952441368/ ------------------------------------------------------------------------ -> Spread the URL: http://www.onelist.com/subscribe/amiga-c <- @ENDNODE @NODE "Mail_55" " Re: 508,338 Search Engines and Directories " From: Allan Odgaard Date: 03-07-00 16:51:56 Subject: @{b}Re: 508,338 Search Engines and Directories@{ub} From: Allan Odgaard On 07-Mar-00, roderick_maclean@agilent.com wrote: > Is there any way that we can prevent spammers from sending > junk to this list? The list manager could modify the settings so that only subscribed members can post, which I think should be done due to the two recent postings. Regards Allan -- URL: http://www.diku.dk/students/duff/ Reality is for people who lack imagination. -> Spread the URL: http://www.onelist.com/subscribe/amiga-c <- @ENDNODE @NODE "Mail_56" " Chinese " From: Boyer Laurent Date: 03-07-00 17:12:52 Subject: @{b}Chinese@{ub} From: "Boyer Laurent" Hello, I wish to start a "Taïwan Chinese" (Big5) Editor. If someone else want to help me or send me helpfull information, i will be the happier man in the world. For the first version i'm only want to display correctly Big5 text. Tanks for all and have a good day. BOYER Laurent (Yes i Know i'm french sorry for my english) lboyer@decan.fr or bl_tyf@club-internet.fr ------------------------------------------------------------------------ DON'T HATE YOUR RATE! Get a NextCard Visa, in 30 seconds! Get rates as low as 0.0% Intro or 9.9% Fixed APR and no hidden fees. Apply NOW! http://click.egroups.com/1/2120/0/_/451227/_/952445658/ ------------------------------------------------------------------------ -> Spread the URL: http://www.onelist.com/subscribe/amiga-c <- @ENDNODE @NODE "Mail_57" " Re: RE: 508,338 Search Engines and Directories " From: "Francis G. Loch" Date: 03-07-00 17:21:41 Subject: @{b}Re: RE: 508,338 Search Engines and Directories@{ub} From: "=?iso-8859-1?q?Francis=20G.=20Loch?=" Hi all, --- roderick_maclean@agilent.com wrote: > > Date: Tue, 07 Mar 2000 02:32:15 -0800 > > From: "508.338 Search Engines and Directories" > > > > Subject: 508,338 Search Engines and Directories > > Is there any way that we can prevent spammers from > sending > junk to this list? > > This is the second piece of spam in two days. The > next thing > they'll try is ripping our email addresses out of > the posts > and I don't really want that to happen - this is my > work > address!! > I totally agree! It's quite annoying to see that I have some mail from this mailing list, only to find out that it's junk mail. > I don't want to unsubscribe from the list either. Nah, I won't be doing that in the near future. ;) Regards, Francis. ===== e-mail:- fgloch@yahoo.com web-site:- http://www.geocities.com/fgloch __________________________________________________ Do You Yahoo!? Talk to your friends online with Yahoo! Messenger. http://im.yahoo.com ------------------------------------------------------------------------ PERFORM CPR ON YOUR APR! Get a NextCard Visa, in 30 seconds! Get rates as low as 0.0% Intro or 9.9% Fixed APR and no hidden fees. Apply NOW! http://click.egroups.com/1/2121/0/_/451227/_/952446115/ ------------------------------------------------------------------------ -> Spread the URL: http://www.onelist.com/subscribe/amiga-c <- @ENDNODE @NODE "Mail_58" " Re: Chinese " From: "Francis G. Loch" Date: 03-07-00 17:28:55 Subject: @{b}Re: Chinese@{ub} From: "=?iso-8859-1?q?Francis=20G.=20Loch?=" Hi there, --- Boyer Laurent wrote: > Hello, > > I wish to start a "Taïwan Chinese" (Big5) Editor. > If someone else want to help me or send me helpfull > information, i will be the happier man in the world. > For the first version i'm only want to display > correctly > Big5 text. I have some Chinese docs in the Hanzi (7-bit encoding?)format, plus a Chinese text viewer to view them. I have no idea if Hanzi is the same, or even similar, to Big5 text. If you want I can send you a copy of this program for your perusal. > Tanks for all and have a good day. Cheers, I will :). Kind regards, Francis. ===== e-mail:- fgloch@yahoo.com web-site:- http://www.geocities.com/fgloch __________________________________________________ Do You Yahoo!? Talk to your friends online with Yahoo! Messenger. http://im.yahoo.com ------------------------------------------------------------------------ GET A NEXTCARD VISA, in 30 seconds! Get rates as low as 0.0% Intro or 9.9% Fixed APR and no hidden fees. Apply NOW! http://click.egroups.com/1/937/0/_/451227/_/952446536/ ------------------------------------------------------------------------ -> Spread the URL: http://www.onelist.com/subscribe/amiga-c <- @ENDNODE @NODE "Mail_59" " Re: RE: 508,338 Search Engines and Directories " From: Andrew Ballard Date: 03-07-00 17:31:59 Subject: @{b}Re: RE: 508,338 Search Engines and Directories@{ub} From: Andrew Ballard > »> Is there any way that we can prevent spammers from »> sending »> junk to this list? »> »> This is the second piece of spam in two days. The »> next thing »> they'll try is ripping our email addresses out of »> the posts »> and I don't really want that to happen - this is my »> work »> address!! »> » »I totally agree! It's quite annoying to see that I »have some mail from this mailing list, only to find »out that it's junk mail. » »> I don't want to unsubscribe from the list either. » »Nah, I won't be doing that in the near future. ;) » I've just been chatting to Mike the list manager, he is actualy having trouble with the list at the moment and can't send or receive any mails to or from it! But he did tell me that he has just ammended the access rights so that only people registered to the list will be able to send mails to it. Hopefully this should filter out any future spam emails. Andy Ballard. andyb@ncc.co.uk ------------------------------------------------------------------------ GET A NEXTCARD VISA, in 30 seconds! Get rates as low as 0.0% Intro APR and no hidden fees. Apply NOW! http://click.egroups.com/1/975/0/_/451227/_/952446758/ ------------------------------------------------------------------------ -> Spread the URL: http://www.onelist.com/subscribe/amiga-c <- @ENDNODE @NODE "Mail_60" " RE: Chinese " From: BOYER Laurent Date: 03-07-00 17:34:15 Subject: @{b}RE: Chinese@{ub} Hi there, too Thank but i already Have it, the Hanzi and the Big5 and not identical the Hanzi is 7-bit and the big5 is two level 16-bits... Kind regards Boyer Laurent -----Original Message----- From: Francis G. Loch [mailto:fgloch@yahoo.com] Sent: mardi 7 mars 2000 17:29 To: amiga-c@onelist.com Subject: Re: [amiga-c] Chinese From: "=?iso-8859-1?q?Francis=20G.=20Loch?=" Hi there, --- Boyer Laurent wrote: > Hello, > > I wish to start a "Taïwan Chinese" (Big5) Editor. > If someone else want to help me or send me helpfull > information, i will be the happier man in the world. > For the first version i'm only want to display > correctly > Big5 text. I have some Chinese docs in the Hanzi (7-bit encoding?)format, plus a Chinese text viewer to view them. I have no idea if Hanzi is the same, or even similar, to Big5 text. If you want I can send you a copy of this program for your perusal. > Tanks for all and have a good day. Cheers, I will :). Kind regards, Francis. ===== e-mail:- fgloch@yahoo.com web-site:- http://www.geocities.com/fgloch __________________________________________________ Do You Yahoo!? Talk to your friends online with Yahoo! Messenger. http://im.yahoo.com _____ Click Here for Move.com! _____ -> Spread the URL: http://www.onelist.com/subscribe/amiga-c <- @ENDNODE @NODE "Mail_61" " RE: Chinese " From: "Francis G. Loch" Date: 03-07-00 18:20:48 Subject: @{b}RE: Chinese@{ub} From: "=?iso-8859-1?q?Francis=20G.=20Loch?=" Hi there, --- BOYER Laurent wrote: > > Hi there, too > > Thank but i already Have it, the Hanzi and the Big5 > and not identical the > Hanzi is 7-bit and the > big5 is two level 16-bits... > Two level 16-bits?! Phew, that's quite a lot (4 bytes per character, if I'm not mistaken). If you're looking for a beta tester for the program I'd be happy to volunteer my services. Kind regards, Francis. ===== e-mail:- fgloch@yahoo.com web-site:- http://www.geocities.com/fgloch __________________________________________________ Do You Yahoo!? Talk to your friends online with Yahoo! Messenger. http://im.yahoo.com ------------------------------------------------------------------------ GET A NEXTCARD VISA, in 30 seconds! Get rates as low as 2.9% Intro or 9.9% Fixed APR and no hidden fees. Apply NOW! http://click.egroups.com/1/936/0/_/451227/_/952449650/ ------------------------------------------------------------------------ -> Spread the URL: http://www.onelist.com/subscribe/amiga-c <- @ENDNODE @NODE "Mail_62" " RE: Chinese " From: BOYER Laurent Date: 03-07-00 18:31:45 Subject: @{b}RE: Chinese@{ub} Hi there, Thank, for you help, but i work all the week and i can only work on it the week-end. i think it will take me a long time to make it run. Do you read traditionnal chinese charactere ? ( About more than 10000 basic ideographic characters... yes only the basic) The big 5 have several level of code: - ascii - chinese char with 1 bytes (The most used) - chinese char with 2 bytes (the currently used) - chinses char with 4 bytes ( the more rarely used) Kind regards, -----Original Message----- From: Francis G. Loch [mailto:fgloch@yahoo.com] Sent: mardi 7 mars 2000 18:21 To: amiga-c@onelist.com Subject: RE: [amiga-c] Chinese From: "=?iso-8859-1?q?Francis=20G.=20Loch?=" Hi there, --- BOYER Laurent wrote: > > Hi there, too > > Thank but i already Have it, the Hanzi and the Big5 > and not identical the > Hanzi is 7-bit and the > big5 is two level 16-bits... > Two level 16-bits?! Phew, that's quite a lot (4 bytes per character, if I'm not mistaken). If you're looking for a beta tester for the program I'd be happy to volunteer my services. Kind regards, Francis. ===== e-mail:- fgloch@yahoo.com web-site:- http://www.geocities.com/fgloch __________________________________________________ Do You Yahoo!? Talk to your friends online with Yahoo! Messenger. http://im.yahoo.com _____ _____ -> Spread the URL: http://www.onelist.com/subscribe/amiga-c <- @ENDNODE @NODE "Mail_63" " BGUI Library " From: Shane O'Neill Date: 03-07-00 21:08:07 Subject: @{b}BGUI Library@{ub} From: Shane O'Neill Hi, I've just started writing a program with the BGUI library, everything was working fine. Though now for some reason the program won't open the library. The strange part is I haven't actually changed my code. I also wrote a small program just to open and close the library (test purposes) though still no luck. Also I've tried replacing the one in the libs directory in case it may have been corrupt. Hasn't helped....Any suggestions? Regards, Shane. ------------------------------------------------------------------------ GET A NEXTCARD VISA, in 30 seconds! Get rates as low as 0.0% Intro APR and no hidden fees. Apply NOW! http://click.egroups.com/1/975/0/_/451227/_/952460885/ ------------------------------------------------------------------------ -> Spread the URL: http://www.onelist.com/subscribe/amiga-c <- @ENDNODE @NODE "Mail_64" " 508,338 Search Engines and Directories " From: "508.338 Search Engines and Directories" Date: 03-07-00 22:58:44 Subject: @{b}508,338 Search Engines and Directories@{ub} From: "508.338 Search Engines and Directories" WORLD SUBMISSION CENTER Blast your site to 508,338+ Search Engines, Directories,Classifieds, Awards Sites, FFA's Free Every Weeks !! http://submit10000.com Submit to 1 FFA from AMEFFA Get a permanent exposure for 3 months Get your ad viewed 300+ times a day GUARANTEE !! http://ameffa.com Become A Reseller For Us And Earn 50% Commission ( 12.50$ ) From Sales Generated >From Your Site, No Investing, Apply Now It's Free !! http://submit10000.com Start Making 400.00$ The First Month Just By Surfing The Net.Double it every 3 Months !! http://submit10000.com ------------------------------------------------------------------------ GET A NEXTCARD VISA, in 30 seconds! Get rates as low as 0.0% Intro or 9.9% Fixed APR and no hidden fees. Apply NOW! http://click.egroups.com/1/937/0/_/451227/_/952455913/ ------------------------------------------------------------------------ -> Spread the URL: http://www.onelist.com/subscribe/amiga-c <- @ENDNODE @NODE "Mail_65" " graphics.library functions under CyberGraphX " From: Gabriele Svelto Date: 03-08-00 09:53:03 Subject: @{b}graphics.library functions under CyberGraphX@{ub} From: Gabriele Svelto Hello everybody, does anybody knows exactly which are the graphics.library functions which don't work properly under CGX or are not at all implemented? The CGX autodocs mention that "almost all functions work under CGX" but I believe that most hardware dependent functions (such as WaitTOF) don't work. Thanx in advance Gabriele ------------------------------------------------------------------------ GET A NEXTCARD VISA, in 30 seconds! Get rates as low as 2.9% Intro or 9.9% Fixed APR and no hidden fees. Apply NOW! http://click.egroups.com/1/936/0/_/451227/_/952505702/ ------------------------------------------------------------------------ -> Spread the URL: http://www.onelist.com/subscribe/amiga-c <- @ENDNODE @NODE "Mail_66" " Re: graphics.library functions under CyberGraphX " From: Michal Bartczak Date: 03-08-00 10:35:05 Subject: @{b}Re: graphics.library functions under CyberGraphX@{ub} From: Michal Bartczak On Wed, 08 Mar 2000, you wrote: > From: Gabriele Svelto > > Hello everybody, > does anybody knows exactly which are the graphics.library functions > which don't work properly under CGX or are not at all implemented? The > CGX autodocs mention that "almost all functions work under CGX" but I > believe that most hardware dependent functions (such as WaitTOF) don't > work. Thanx in advance > > Gabriele AFIR WaitBOVP() works nice under some cards, under some else it does nothing. But don't expect cgx to analyse your custom copperlists ;))) Michal ------------------------------------------------------------------------ GET A NEXTCARD VISA, in 30 seconds! Get rates as low as 2.9% Intro or 9.9% Fixed APR and no hidden fees. Apply NOW! http://click.egroups.com/1/936/0/_/451227/_/952507827/ ------------------------------------------------------------------------ -> Spread the URL: http://www.onelist.com/subscribe/amiga-c <- @ENDNODE @NODE "Mail_67" " Does anybody know how to program Linux serial ports? " From: Simon John Hall Date: 03-08-00 11:42:14 Subject: @{b}Does anybody know how to program Linux serial ports?@{ub} From: Simon John Hall I have ported a lot of Linux code but now i have to convert a lot of Linux code to Amiga code and i haven't the faintest idea when it comes to understanding the different arguments. any help from anybody? if it works then we'll have support for 100+ digital cameras! Simon Hall ------------------------------------------------------------------------ DON'T HATE YOUR RATE! Get a NextCard Visa, in 30 seconds! Get rates as low as 0.0% Intro or 9.9% Fixed APR and no hidden fees. Apply NOW! http://click.egroups.com/1/2120/0/_/451227/_/952512157/ ------------------------------------------------------------------------ -> Spread the URL: http://www.onelist.com/subscribe/amiga-c <- @ENDNODE @NODE "Mail_68" " Re: BGUI Library " From: "Francis G. Loch" Date: 03-08-00 11:59:22 Subject: @{b}Re: BGUI Library@{ub} From: "=?iso-8859-1?q?Francis=20G.=20Loch?=" Hi Shane, --- Shane O'Neill wrote: > Hi, > > I've just started writing a program with the BGUI > library, everything was > working fine. Though now for some reason the program > won't open the library. > The strange part is I haven't actually changed my > code. > I also wrote a small program just to open and close > the library (test > purposes) though still no luck. > Also I've tried replacing the one in the libs > directory in case it may have > been corrupt. Hasn't helped....Any suggestions? I would try re-installing the whole of the BGUI library, not just the BGUI.library file. I would also try checking the version number of the library. Kind regards, Francis. ===== e-mail:- fgloch@yahoo.com web-site:- http://www.geocities.com/fgloch __________________________________________________ Do You Yahoo!? Talk to your friends online with Yahoo! Messenger. http://im.yahoo.com ------------------------------------------------------------------------ MAXIMIZE YOUR CARD, MINIMIZE YOUR RATE! Get a NextCard Visa, in 30 seconds! Get rates as low as 0.0% Intro or 9.9% Fixed APR and no hidden fees. Apply NOW! http://click.egroups.com/1/2122/0/_/451227/_/952513163/ ------------------------------------------------------------------------ -> Spread the URL: http://www.onelist.com/subscribe/amiga-c <- @ENDNODE @NODE "Mail_69" " Re: graphics.library functions under CyberGraphX " From: Emanuele Cesaroni Date: 03-08-00 13:35:57 Subject: @{b}Re: graphics.library functions under CyberGraphX@{ub} From: Emanuele Cesaroni Hello Gabriele On 08-Mar-00, you wrote: > From: Gabriele Svelto > > Hello everybody, > does anybody knows exactly which are the graphics.library functions > which don't work properly under CGX or are not at all implemented? The > CGX autodocs mention that "almost all functions work under CGX" but I > believe that most hardware dependent functions (such as WaitTOF) don't > work. Thanx in advance > > Gabriele > Hello Gabriele. I am not sure about all the graphics.library functions, so i don't know if WaitTOF works. WaitBOVP works perfectly both on Cgx and Picasso96 for example. I am sure only about all the graphics functions which use the blitters on graphics boards which could give you a lot of problems. On Picasso96 all the minterms of them, also the more exotic, are implemented well, while on cgx are not. On cgx you can't use BltMaskBitMapRastPort() inverting the mask, in all the depth formats (8,16,24) the minterm is $ca or $ac, i don't remember well, so you need to do two masks before blit, one normal and the other the logical not of the first, or invert the first each times by Cpu. Moreover you can't do on 16 and 24 bit bitmaps a copy inverting the source, using BltBitMap(). On 8 bit depth bitmaps works fine. Probably the same for BltBitMapRastPort() and ClipBlit(), i never try. Bye. Emanuele. ------------------------------------------------------------------------ DON'T HATE YOUR RATE! Get a NextCard Visa, in 30 seconds! Get rates as low as 0.0% Intro or 9.9% Fixed APR and no hidden fees. Apply NOW! http://click.egroups.com/1/2120/0/_/451227/_/952554053/ ------------------------------------------------------------------------ @ENDNODE @NODE "Mail_70" " Re: Does anybody know how to program Linux serial ports? " From: Olaf Barthel Date: 03-08-00 14:51:06 Subject: @{b}Re: Does anybody know how to program Linux serial ports?@{ub} From: "Olaf Barthel" On Mar 8 Simon (Simon John Hall) wrote: > From: Simon John Hall > > I have ported a lot of Linux code but now i have to convert a lot of > Linux code to Amiga code and i haven't the faintest idea when it comes > to understanding the different arguments. any help from anybody? if it > works then we'll have support for 100+ digital cameras! There's a HOWTO document which covers this problem. You might also want to dig into your pocket and buy the book "Advanced programming in the Unix environment" by W. Richard Stevens. It explains the issues, provides examples and covers practically everything else you would ever want to know about programming a Unix or Unix-like machine. -- Home: Olaf Barthel, Brabeckstrasse 35, D-30559 Hannover Net: olsen@sourcery.han.de (Home), olsen@logicalline.com (Work) ------------------------------------------------------------------------ DON'T HATE YOUR RATE! Get a NextCard Visa, in 30 seconds! Get rates as low as 0.0% Intro or 9.9% Fixed APR and no hidden fees. Apply NOW! http://click.egroups.com/1/2120/0/_/451227/_/952542834/ ------------------------------------------------------------------------ @ENDNODE @NODE "Mail_71" " RE: graphics.library functions under CyberGraphX " From: Jesper Svennevid Date: 03-08-00 18:51:27 Subject: @{b}RE: graphics.library functions under CyberGraphX@{ub} From: Jesper Svennevid > From: Gabriele Svelto > > Hello everybody, > does anybody knows exactly which are the graphics.library functions > which don't work properly under CGX or are not at all implemented? The > CGX autodocs mention that "almost all functions work under CGX" but I > believe that most hardware dependent functions (such as WaitTOF) don't > work. Thanx in advance WaitTOF() works, atleast in later revisions of CGFX, due to that there are retrace(when the display enters the vertical blanking period) on gfx-cards aswell. -- // Jesper Svennevid : jesper@fxrealm.com Software Developer at FXRealm Studios http://www.fxrealm.com/ ------------------------------------------------------------------------ GET A NEXTCARD VISA, in 30 seconds! Get rates as low as 0.0% Intro APR and no hidden fees. Apply NOW! http://click.egroups.com/1/975/0/_/451227/_/952537905/ ------------------------------------------------------------------------ @ENDNODE @NODE "Mail_72" " RE: graphics.library functions under CyberGraphX " From: Paul Hill Date: 03-08-00 21:46:06 Subject: @{b}RE: graphics.library functions under CyberGraphX@{ub} From: "Paul Hill" On 08-Mar-00 17:51:27, Jesper Svennevid wrote: >From: Jesper Svennevid >> From: Gabriele Svelto >> >> Hello everybody, >> does anybody knows exactly which are the graphics.library functions >> which don't work properly under CGX or are not at all implemented? The >> CGX autodocs mention that "almost all functions work under CGX" but I >> believe that most hardware dependent functions (such as WaitTOF) don't >> work. Thanx in advance >WaitTOF() works, atleast in later revisions of CGFX, due to that there >are retrace(when the display enters the vertical blanking period) on >gfx-cards aswell. I'm pretty sure that WaitTOF() still waits for the TOF of the native ECS/AGA display. Perhaps I'm wrong though. -- Paul ------------------------------------------------------------------------ GET A NEXTCARD VISA, in 30 seconds! Get rates as low as 2.9% Intro or 9.9% Fixed APR and no hidden fees. Apply NOW! http://click.egroups.com/1/936/0/_/451227/_/952549915/ ------------------------------------------------------------------------ @ENDNODE @NODE "Mail_73" " pointer array " From: Johan Rasten Date: 03-08-00 22:15:26 Subject: @{b}pointer array@{ub} From: Johan Rasten Hi again! I'm having a little trouble creating a pointer array, of variable size, inside a class.. It ought to be something like this: (I think atleast) class blabla { /public/: blabla(UWORD size); APTR get_pointer(UWORD number); /private/: APTR pointers; } blabla::blabla(UWORD size) { pointers = new APTR[size]; } APTR blabla::get_pointer(UWORD number) { return pointer[number]; } But it doesn't work ;,( I don't know exactly what the compiler tries to do, but the elements of pointer[] seems to be something elsa than APTRs atleast! Please help fast! Live long and Prosper /alt3reg0 ps. Can anybody please explain the *inline* and *virtual* keywords?? -- http://hem1.passagen.se/zman/main.html The only job where you start at the top is digging a hole. ------------------------------------------------------------------------ GET A NEXTCARD VISA, in 30 seconds! Get rates as low as 0.0% Intro or 9.9% Fixed APR and no hidden fees. Apply NOW! http://click.egroups.com/1/937/0/_/451227/_/952550544/ ------------------------------------------------------------------------ @ENDNODE @NODE "Mail_74" " Re: pointer array " From: Niels de Koning Date: 03-08-00 22:52:14 Subject: @{b}Re: pointer array@{ub} From: Niels de Koning [...] > /private/: > APTR pointers; [...] I don't know exactly what you meant when you described your problem with this code. Do you mean it does not compile or do you mean it gives strange results? From what I see hear I suppose it does not compile. I have not verified if this is true by trying to compile the code myself, but I think your problem lies in the fact that the private variable in the class is of the type APTR and NOT of the type 'pointer to an array of type APTR', which is what you ask for with the 'new' operator. Hope this helps.... Regards, // Niels de Koning ____________________________________________________________________ Co-editor on 3DAddict's GFXZone http://www.gfxzone.org ------------------------------------------------------------------------ DON'T HATE YOUR RATE! Get a NextCard Visa, in 30 seconds! Get rates as low as 0.0% Intro or 9.9% Fixed APR and no hidden fees. Apply NOW! http://click.egroups.com/1/2120/0/_/451227/_/952552364/ ------------------------------------------------------------------------ @ENDNODE @NODE "Mail_75" " Re: pointer array " From: Andreas Meyer Date: 03-09-00 11:40:59 Subject: @{b}Re: pointer array@{ub} From: Andreas Meyer Hello Johan On 08-Mar-00, Johan Rasten wrote: > From: Johan Rasten > > Hi again! I'm having a little trouble creating a pointer array, of > variable size, inside a class.. It ought to be something like this: (I > think atleast) > > class blabla { > /public/: > blabla(UWORD size); > APTR get_pointer(UWORD number); > > /private/: > APTR pointers; > } should be: private: APTR *pointers; > > blabla::blabla(UWORD size) > { > pointers = new APTR[size]; > } > > APTR blabla::get_pointer(UWORD number) > { > return pointer[number]; > } > OK, just a typo: return pointers[number]; > But it doesn't work ;,( I don't know exactly what the compiler tries to > do, but the elements of pointer[] seems to be something elsa than APTRs > atleast! Please help fast! "pointers" needs to be declared as a pointer to APTR, so the compiler knows the scope of the pointer and pointers[x] returns an APTR > Live long and Prosper > > /alt3reg0 > > ps. Can anybody please explain the *inline* and *virtual* keywords?? uhh, "inline" is meant for inlining the function code :-) normally a function (method) call results in leaving the main stream of your function, branching to the functions code and then returning to the code after you function call. when a function is inlined (preferrable a short piece of code), a copy of the code that is meant to be in the function is placed everywhere where the function gets called, so it is faster. the main stream of your code is not interrupted by branches. "virtual" is used with dynamic binding and polymorphisms. unless you use something like a pointer to a base class object to build dynamically objects of derived classes of that base class (sounds difficult, hmm ?), you will not have to use it. In that cases "virtual" gives you a way to get functions pointers, pointing to the base class function or the derived class function dynamically. I don^t know if this helps, maybe you better get some documentation. The Coronado Enterprises C++ Tutorial is free and online avilable. It comes with lots of source. A Ansi-C tutorial is also there. Good luck, Andreas -- Our sails swell full as we brave all seas On a westward wind to live as we please ------------------------------------------------------------------------ GET A NEXTCARD VISA, in 30 seconds! Get rates as low as 0.0% Intro or 9.9% Fixed APR and no hidden fees. Apply NOW! http://click.egroups.com/1/937/0/_/451227/_/952602229/ ------------------------------------------------------------------------ @ENDNODE @NODE "Mail_76" " Chinese text editor " From: BOYER Laurent Date: 03-09-00 13:50:05 Subject: @{b}Chinese text editor@{ub} Hello, I will start the translation of the "ChinesePower 3.0" wordprocessor from the x-windows version, and i wish to know where i can found more data about the port from unix-x11 (ixemul.library). this work is based on "Chinesepower 3.0" for text display and xcterm 5.0 for text input. If you want the source code go to the "ifcss" ftp site. Thanks, Have a good day... BOYER Laurent @ENDNODE @NODE "Mail_77" " Creating ARexx variables " From: Richard Drummond Date: 03-09-00 15:17:48 Subject: @{b}Creating ARexx variables@{ub} From: "Richard Drummond" Hello everybody. This is question to do with making a C program act as an ARexx host. I want to be able to return the results of an ARexx command in a named variable, possibly a stemmed variable. How do I create create such a variable from C? Does anybody have some example code or can anybody point me in the right direction? Cheers, Rich -- Richard Drummond Staff Writer, Amiga Format mailto: richard.drummond@futurenet.co.uk pgp : http://www.drummond.u-net.com/download/richards_key.asc phone : +44 (0)1225 442244 ext 2417 ------------------------------------------------------------------------ DON'T HATE YOUR RATE! Get a NextCard Visa, in 30 seconds! Get rates as low as 0.0% Intro or 9.9% Fixed APR and no hidden fees. Apply NOW! http://click.egroups.com/1/2120/0/_/451227/_/952615333/ ------------------------------------------------------------------------ @ENDNODE @NODE "Mail_78" " RE: Chinese " From: "Francis G. Loch" Date: 03-09-00 18:12:17 Subject: @{b}RE: Chinese@{ub} From: "=?iso-8859-1?q?Francis=20G.=20Loch?=" Hi Boyer, --- BOYER Laurent wrote: > Hi there, > > Thank, for you help, but i work all the week and > i can only work on it > the week-end. > i think it will take me a long time to make it > run. > Do you read traditionnal chinese charactere ? A little. I have a lot of pop music which is in Cantonese so my interest in learning to read Chinese (and speak Cantonese) stemmed from that. > ( About more than 10000 basic ideographic > characters... yes only the > basic) > The big 5 have several level of code: > - ascii > - chinese char with 1 bytes (The most used) > - chinese char with 2 bytes (the currently used) > - chinses char with 4 bytes ( the more rarely > used) > Kind regards, Francis. ===== e-mail:- fgloch@yahoo.com web-site:- http://www.geocities.com/fgloch __________________________________________________ Do You Yahoo!? Talk to your friends online with Yahoo! Messenger. http://im.yahoo.com ------------------------------------------------------------------------ MAXIMIZE YOUR CARD, MINIMIZE YOUR RATE! Get a NextCard Visa, in 30 seconds! Get rates as low as 0.0% Intro or 9.9% Fixed APR and no hidden fees. Apply NOW! http://click.egroups.com/1/2122/0/_/451227/_/952622421/ ------------------------------------------------------------------------ @ENDNODE @NODE "Mail_79" " 16bit signed to binary " From: Charles Barr Date: 03-09-00 19:42:00 Subject: @{b}16bit signed to binary@{ub} From: Charles Barr Hello, I have a friend whos, wants to know how you would convert a signed 16bit number to into a binary pattern eg. 2 = 00000000000010 etc. Any help would br appreciated Cheers -- Among the detectives and patrol officers crowded over the body of Latonya Kim Wallace there is no easy banter, no course exchange of cop humor or time-worn indifference. -- David Simon ------------------------------------------------------------------------ DON'T HATE YOUR RATE! Get a NextCard Visa, in 30 seconds! Get rates as low as 0.0% Intro or 9.9% Fixed APR and no hidden fees. Apply NOW! http://click.egroups.com/1/2120/0/_/451227/_/952627210/ ------------------------------------------------------------------------ @ENDNODE @NODE "Mail_80" " Re: 16bit signed to binary " From: Tim Hanson Date: 03-09-00 20:17:30 Subject: @{b}Re: 16bit signed to binary@{ub} From: Tim Hanson Hi Charles Charles Barr wrote this on 09-Mar-00 at 19:42:00 > From: Charles Barr > > Hello, > I have a friend whos, wants to know how you would convert a > signed 16bit number to into a binary pattern > > eg. > > 2 = 00000000000010 etc. Theres no easy way I`m afraid. You just ahve to do it manually. If you want some code give us a shout. -- Tim Hanson, by the River,Canterbury, Kent. Amiga A1200T,`040/25,603e/160,32MB Fast,BVisionPPC, 8.4GB IDE HD,ZIP SCSI,RENO x2CD SCSI,Pace 56K Don`t waste your time by visiting http://www.sinister67.freeserve.co.uk ------------------------------------------------------------------------ GET A NEXTCARD VISA, in 30 seconds! Get rates as low as 0.0% Intro or 9.9% Fixed APR and no hidden fees. Apply NOW! http://click.egroups.com/1/937/0/_/451227/_/952635293/ ------------------------------------------------------------------------ @ENDNODE @NODE "Mail_81" " Re: pointer array " From: Gabriele Svelto Date: 03-09-00 20:43:35 Subject: @{b}Re: pointer array@{ub} From: Gabriele Svelto Hi Johan On 08-Mar-00, you wrote: > From: Johan Rasten > > Hi again! I'm having a little trouble creating a pointer array, of > variable size, inside a class.. It ought to be something like this: (I > think atleast) > > class blabla { > /public/: > blabla(UWORD size); > APTR get_pointer(UWORD number); > > /private/: > APTR pointers; > } > > blabla::blabla(UWORD size) > { > pointers = new APTR[size]; > } > > APTR blabla::get_pointer(UWORD number) > { > return pointer[number]; > } > > But it doesn't work ;,( I don't know exactly what the compiler tries > to do, but the elements of pointer[] seems to be something elsa than > APTRs atleast! Please help fast! This could be explained by the fact that you create the array dynamically and access to it like an array, try return pointer + number; instead of return pointer[number] (unless you have overloaded []) > Live long and Prosper > > /alt3reg0 > > ps. Can anybody please explain the *inline* and *virtual* keywords?? inline tells the compiler not to call the function but to inline it like it was a macro so for example: inline int square(int num) { return num * num; } works pretty much like #define SQUARE(num) ((num) * (num)) but with type-checking of the operands (thus it's safer to use than a macro), virtual (if I don't go wrong) tells a class that the functions defined inside it can be redefined in derived classes, for example: virtual class bla { public: int DoSomething( void ); private: int something; } int bla::DoSomething( void ) { /* code here */ } class blabla : class bla { /* ... */ } int blabla::DoSomething( void ) { /* different code */ } Thus you can redefin already defined functions in derived classes of a virtual class. See ya Gabriele ------------------------------------------------------------------------ GET A NEXTCARD VISA, in 30 seconds! Get rates as low as 0.0% Intro or 9.9% Fixed APR and no hidden fees. Apply NOW! http://click.egroups.com/1/937/0/_/451227/_/952631086/ ------------------------------------------------------------------------ @ENDNODE @NODE "Mail_82" " Re: pointer array " From: Johan Rasten Date: 03-09-00 21:20:23 Subject: @{b}Re: pointer array@{ub} From: Johan Rasten Thanks all! I've got it up and working now (damn, my game is going to be beautiful! ;-P). Does anybody have a good (working for beta testing) map generator? Just something that creates a VAR_X * VAR_Y sized map with some land, sea, mountains. The "problem" (I figure I could make one myself, but it would take a few days of precious coding time as I'm such a perfectionist ;-PP) is that the top/bottom and left/right edges should match so you can travel around the world.. Now I also have a new problem: When I tell the display class to open a weird sized display, it matches the best resolution and opens that screen with a window with the requested size instead. It all works fine until I switch to the second buffer, where the area outside the window is filled with junk (the c2p buffer usually). How can I avoid this? I've ripped most of the double buffer code (from the sys 5 I think) so I don't understand all of it. Please help! (again) RegardZ /J-J-J-Johan ps. I didn't understand that virtual stuff, but I don't think I have to use it, well structured OO coding isn't what I'm best at ;-P And someone ought to write a kick-ass amiga emulator for mac, so I can get an iMac and run AmigaOS/MorhOS/P.os or whatever.. macos sucks -- http://hem1.passagen.se/zman/main.html Pretend to spank me - I'm a pseudo-masochist! ------------------------------------------------------------------------ PERFORM CPR ON YOUR APR! Get a NextCard Visa, in 30 seconds! Get rates as low as 0.0% Intro or 9.9% Fixed APR and no hidden fees. Apply NOW! http://click.egroups.com/1/2121/0/_/451227/_/952633632/ ------------------------------------------------------------------------ @ENDNODE @NODE "Mail_83" " RE: 16bit signed to binary " From: Jesper Svennevid Date: 03-09-00 22:08:39 Subject: @{b}RE: 16bit signed to binary@{ub} From: Jesper Svennevid > -----Original Message----- > From: Charles Barr [mailto:charlie@amigaforever.free-online.co.uk] > Sent: Thursday, March 09, 2000 7:42 PM > To: AmigaC > Subject: [amiga-c] 16bit signed to binary > > > From: Charles Barr > > Hello, > I have a friend whos, wants to know how you would convert a > signed 16bit number to into a binary pattern > > eg. > > 2 = 00000000000010 etc. > > > Any help would br appreciated Hm, this should do the trick: void short_to_binary(short value, char* output) // output should be atleast 17 chars long { int i; for(i=32768;i>0;i>>=1) *output++ = (value&i)?'1':'0'; *output = '\0'; // terminate string } Explanation: Loops through the value from bit 15 to bit 0, testing if it's set, and then writes that status(1/0) to the output-string. -- // Jesper Svennevid : jesper@fxrealm.com Software Developer at FXRealm Studios http://www.fxrealm.com/ @ENDNODE @NODE "Mail_84" " Re: 16bit signed to binary " From: Robin Cloutman Date: 03-09-00 22:13:44 Subject: @{b}Re: 16bit signed to binary@{ub} From: Robin Cloutman On Thursday, 09-Mar-00, Jesper wrote: > From: Jesper Svennevid > >> -----Original Message----- >> From: Charles Barr [mailto:charlie@amigaforever.free-online.co.uk] >> Sent: Thursday, March 09, 2000 7:42 PM >> To: AmigaC >> Subject: [amiga-c] 16bit signed to binary >> >> >> From: Charles Barr >> >> Hello, >> I have a friend whos, wants to know how you would convert a >> signed 16bit number to into a binary pattern >> >> eg. >> >> 2 = 00000000000010 etc. >> >> >> Any help would br appreciated > > Hm, this should do the trick: > > void short_to_binary(short value, char* output) // output should be > atleast 17 chars long > { > int i; > > for(i=32768;i>0;i>>=1) *output++ = (value&i)?'1':'0'; > *output = '\0'; // terminate string > } > > Explanation: > > Loops through the value from bit 15 to bit 0, testing if it's set, and > then writes that status(1/0) to the output-string. > Slightly nicer (imho) code - void short_to_binary( short value, char *output ) { int i; for ( i = 0 ; i < 16 i++ )output[i] = (value&(1< This is question to do with making a C program act as an ARexx host. I RD> want to be able to return the results of an ARexx command in a named RD> variable, possibly a stemmed variable. How do I create create such a RD> variable from C? AFAIK, you can only use SetRexxVar() from Amiga.lib (or something based on the same idea/code). You can't use the result variables for it. -- Neil Williams, British. mailto:neil@aio.co.uk ICQ:18223711 2:442/107 AIO: http://www.aio.co.uk TKG: http://www.tkgbbs.freeserve.co.uk Zeus Developments, BBS s/w: http://www.bleach.demon.co.uk/zeus/ ------------------------------------------------------------------------ PERFORM CPR ON YOUR APR! Get a NextCard Visa, in 30 seconds! Get rates as low as 0.0% Intro or 9.9% Fixed APR and no hidden fees. Apply NOW! http://click.egroups.com/1/2121/0/_/451227/_/952712375/ ------------------------------------------------------------------------ @ENDNODE @NODE "Mail_86" " Re: Re: BGUI Library " From: Shane O'Neill Date: 03-10-00 02:29:31 Subject: @{b}Re: Re: BGUI Library@{ub} From: Shane O'Neill >From: Francis G. Loch >I would try re-installing the whole of the BGUI >library, not just the BGUI.library file. >I would also try checking the version number of the >library. Cheers. Hmmm, my first response didn't seem to make it to the list. Anyway, I have it sorted out now. I't was a bit stupid really. The bgui.library from the developers archive did not work, and when I placed it in libs: I forgot to remove version 41.8 from execs library list, so next time I restarted I couldn't use any BGUI stuff. Oops. Regards, Shane. ------------------------------------------------------------------------ PERFORM CPR ON YOUR APR! Get a NextCard Visa, in 30 seconds! Get rates as low as 0.0% Intro or 9.9% Fixed APR and no hidden fees. Apply NOW! http://click.egroups.com/1/2121/0/_/451227/_/952651853/ ------------------------------------------------------------------------ @ENDNODE @NODE "Mail_87" " Re: 16bit signed to binary " From: Colin Wenzel Date: 03-10-00 04:57:02 Subject: @{b}Re: 16bit signed to binary@{ub} From: Colin Wenzel On 10-Mar-00, Charles Barr wrote: > From: Charles Barr > Hello, > I have a friend whos, wants to know how you would convert a > signed 16bit number to into a binary pattern > eg. > > 2 = 00000000000010 etc. > > Any help would br appreciated > > Cheers Here you go...... This one does all 32 bits, so if you want only the low 16 bits, just add 16 to the return pointer value... ------------------------------------- Colin Wenzel. Australia. EMAIL: mailto:colstv@hotkey.net.au URL: http://www.hotkey.net.au/~colstv/ ICQ: 17608330 AMIGA: 4000T, 060/50, 150Mb RAM, O.S 3.5 --------------------------------------- ------------------------------------------------------------------------ MAXIMIZE YOUR CARD, MINIMIZE YOUR RATE! Get a NextCard Visa, in 30 seconds! Get rates as low as 0.0% Intro or 9.9% Fixed APR and no hidden fees. Apply NOW! http://click.egroups.com/1/2122/0/_/451227/_/952660769/ ------------------------------------------------------------------------ @ENDNODE @NODE "Mail_88" " Re: Re: pointer array " From: Gregory Roberts II Date: 03-10-00 05:39:50 Subject: @{b}Re: Re: pointer array@{ub} From: Gregory Roberts II Hi Johan, On Thu, 9 Mar 2000, Johan Rasten wrote: > Thanks all! I've got it up and working now (damn, my game is going to be > beautiful! ;-P). Does anybody have a good (working for beta testing) map > generator? Just something that creates a VAR_X * VAR_Y sized map with some I have just the thing you are looking for, except for one little problem. It's written in assembler and it's been a while since I last looked at it. If you're not in a big hurry I might be able to convert it to some reasonable looking 'C' code, or at least explain how it works. :-) It uses a pseudo-random number generator to fractally divide the map area and adjust the 'altitude' up or down relative to the original. It does this to so many levels, then averages the area around each location so that it will pick up some of the surronding terrain and make smooth transitions in elevation. It generates some beautiful maps, if I do say so myself. :-) The beauty of it is, you can generate a near infinite number of maps (well, 4 billion or so anyway :-), but if you start with the same seed, you always get the same map, so your program only has to store a 32-bit number to have it's complete map! > land, sea, mountains. The "problem" (I figure I could make one myself, but > it would take a few days of precious coding time as I'm such a perfectionist > ;-PP) is that the top/bottom and left/right edges should match so you can > travel around the world.. > This is actually pretty easy to achieve. Basically all you have to do is duplicate part of each edge of the map on the other side so you have it around to display until you actually reach the edge. If you really want a drawn out explanation of this, just let me know. :-) The array indexing can get pretty hairy. Just try it with a hexagonal map! (Yes, it can be done ;) Anyway, if I get a chance, I'll try to post something. Feel free to "remind" me about it if I'm slow. Greg -- gregr@en.com KC8BGC Amiga Enthusiast :) IBM programmer :( ------------------------------------------------------------------------ GET A NEXTCARD VISA, in 30 seconds! Get rates as low as 0.0% Intro or 9.9% Fixed APR and no hidden fees. Apply NOW! http://click.egroups.com/1/937/0/_/451227/_/952663314/ ------------------------------------------------------------------------ @ENDNODE @NODE "Mail_89" " RE: Re: 16bit signed to binary " From: Jesper Svennevid Date: 03-10-00 12:30:11 Subject: @{b}RE: Re: 16bit signed to binary@{ub} From: Jesper Svennevid > From: Robin Cloutman > > On Thursday, 09-Mar-00, Jesper wrote: > > From: Jesper Svennevid > > Hm, this should do the trick: > > > > void short_to_binary(short value, char* output) // output should be > > atleast 17 chars long > > { > > int i; > > > > for(i=32768;i>0;i>>=1) *output++ = (value&i)?'1':'0'; > > *output = '\0'; // terminate string > > } > > > > Explanation: > > > > Loops through the value from bit 15 to bit 0, testing if > it's set, and > > then writes that status(1/0) to the output-string. > > > > Slightly nicer (imho) code - > > void short_to_binary( short value, char *output ) > { > int i; > > for ( i = 0 ; i < 16 i++ )output[i] = (value&(1< output[i] = '\0'; > } yep, but not faster atleast :) as you have an additinal add (i++), a pointer offset (a incrementing pointer access is faster),and that the routine in question doesn't work as it should. :) missing ; in the for-loop, " instead of ' surrounding the characters, and the entire routine produces the entire string backwards. :) But that's just simple errors, as I assume you wrote the routine directly in the mail. :) This should be the fixed version of your routine: void short_to_binary(short value, char* output) { int i; for(i = 0; i<16;i++) output[i] = (value&(1<<(15-i)))?'1':'0'; output[i] = '\0'; } And I must say that I find my routine simpler :) although, to simplify understanding of the routine, it could be written like this: void short_to_binary(short value, char* output) { int i; for(i= 1 << (16 - 1) ; i > 0; i >>= 1) *(output++) = (value & i) ? '1' : '0'; *output = '\0'; } -- // Jesper Svennevid : jesper@fxrealm.com Software Developer at FXRealm Studios http://www.fxrealm.com/ ------------------------------------------------------------------------ GET A NEXTCARD VISA, in 30 seconds! Get rates as low as 0.0% Intro or 9.9% Fixed APR and no hidden fees. Apply NOW! http://click.egroups.com/1/937/0/_/451227/_/952691494/ ------------------------------------------------------------------------ @ENDNODE @NODE "Mail_90" " Reaction " From: mike.carter@ukonline.co.uk Date: 03-10-00 13:13:12 Subject: @{b}Reaction@{ub} From: mike.carter@ukonline.co.uk Hi, I don't own OS3.5, nor the new developer CD (yet) but would like to know how you go about creating simple GUIs in Reaction (without using the GUI creator). Does anybody have a small piece of code with a GUI in it? Or does it use that external file thingy stormwizard.library or something? Anyway I'm just curious for now... THanks, M!KE C. ------------------------------------------------------------------------ GET A NEXTCARD VISA, in 30 seconds! Get rates as low as 2.9% Intro or 9.9% Fixed APR and no hidden fees. Apply NOW! http://click.egroups.com/1/936/0/_/451227/_/952690395/ ------------------------------------------------------------------------ @ENDNODE @NODE "Mail_91" " 14 bit audio " From: Paul Hill Date: 03-10-00 16:51:07 Subject: @{b}14 bit audio@{ub} From: Paul Hill Hi All, Does anyone know how those 14bit audio tricks work? Is it really just a matter of playing the most significant byte at full volume and the least significant at half volume? -- Paul ------------------------------------------------------------------------ DON'T HATE YOUR RATE! Get a NextCard Visa, in 30 seconds! Get rates as low as 0.0% Intro or 9.9% Fixed APR and no hidden fees. Apply NOW! http://click.egroups.com/1/2120/0/_/451227/_/952703669/ ------------------------------------------------------------------------ @ENDNODE @NODE "Mail_92" " Re: 14 bit audio " From: Paul Hill Date: 03-10-00 17:30:36 Subject: @{b}Re: 14 bit audio@{ub} From: Paul Hill Paul Hill wrote: > > From: Paul Hill > > Hi All, > > Does anyone know how those 14bit audio tricks work? Is it really just a > matter of playing the most significant byte at full volume and the least > significant at half volume? Hi Paul! Er, actually I found the info myself. Apparently the second channel should be played at 1/64th of the volume. -- Paul ------------------------------------------------------------------------ GET A NEXTCARD VISA, in 30 seconds! Get rates as low as 0.0% Intro or 9.9% Fixed APR and no hidden fees. Apply NOW! http://click.egroups.com/1/937/0/_/451227/_/952705909/ ------------------------------------------------------------------------ @ENDNODE @NODE "Mail_93" " Re: 14 bit audio " From: "Francis G. Loch" Date: 03-10-00 18:10:58 Subject: @{b}Re: 14 bit audio@{ub} From: "=?iso-8859-1?q?Francis=20G.=20Loch?=" Hi there, --- Paul Hill wrote: > Paul Hill wrote: > > > > From: Paul Hill > > > > Hi All, > > > > Does anyone know how those 14bit audio tricks > work? Is it really just a > > matter of playing the most significant byte at > full volume and the least > > significant at half volume? > > Hi Paul! Er, actually I found the info myself. > Apparently the second channel > should be played at 1/64th of the volume. > > -- > Paul > I might be mistaken, but I thought that the 14-bit audio worked because the Paula chip's "sound department" is actually 16-bits. Normally 8-bits are used for the sample and the other 8-bits are used for things like the volume, etc.. The 14-bit audio routines work by utilising 6 of the other 8-bits that weren't essential for producing audio. Anyone feel free to correct me if I'm wrong. It's the only way that I'll learn ;). Kind regards, Francis. ===== e-mail:- fgloch@yahoo.com web-site:- http://www.geocities.com/fgloch __________________________________________________ Do You Yahoo!? Talk to your friends online with Yahoo! Messenger. http://im.yahoo.com ------------------------------------------------------------------------ PERFORM CPR ON YOUR APR! Get a NextCard Visa, in 30 seconds! Get rates as low as 0.0% Intro or 9.9% Fixed APR and no hidden fees. Apply NOW! http://click.egroups.com/1/2121/0/_/451227/_/952708268/ ------------------------------------------------------------------------ @ENDNODE @NODE "Mail_94" " Re: 16bit signed to binary " From: Robin Cloutman Date: 03-10-00 19:16:57 Subject: @{b}Re: 16bit signed to binary@{ub} From: Robin Cloutman On Friday, 10-Mar-00, Jesper wrote: > From: Jesper Svennevid > >> From: Robin Cloutman >> >> On Thursday, 09-Mar-00, Jesper wrote: >>> From: Jesper Svennevid >>> Hm, this should do the trick: >>> >>> void short_to_binary(short value, char* output) // output should be >>> atleast 17 chars long >>> { >>> int i; >>> >>> for(i=32768;i>0;i>>=1) *output++ = (value&i)?'1':'0'; >>> *output = '\0'; // terminate string >>> } >>> >>> Explanation: >>> >>> Loops through the value from bit 15 to bit 0, testing if >> it's set, and >>> then writes that status(1/0) to the output-string. >>> >> >> Slightly nicer (imho) code - >> >> void short_to_binary( short value, char *output ) >> { >> int i; >> >> for ( i = 0 ; i < 16 i++ )output[i] = (value&(1<> output[i] = '\0'; >> } > > yep, but not faster atleast :) as you have an additinal add (i++), > a pointer offset (a incrementing pointer access is faster),and that > the routine in question doesn't work as it should. :) > > missing ; in the for-loop, " instead of ' surrounding the characters, > and the entire routine produces the entire string backwards. :) But > that's just simple errors, as I assume you wrote the routine directly > in the mail. :) doh! yeah, i did write it in the mail and not having my normal text editor... ;-) > This should be the fixed version of your routine: > > void short_to_binary(short value, char* output) > { > int i; > > for(i = 0; i<16;i++) output[i] = (value&(1<<(15-i)))?'1':'0'; > output[i] = '\0'; > } > > And I must say that I find my routine simpler :) although, to > simplify understanding of the routine, it could be written like this: > > void short_to_binary(short value, char* output) > { > int i; > > for(i= 1 << (16 - 1) ; i > 0; i >>= 1) > *(output++) = (value & i) ? '1' : '0'; > > *output = '\0'; > } Yeah, this is faster, but my routine uses barely any longer, proper opomising on the compiler end should result in 1 opcode extra, my routine is easier to read (I tend to use an IS_SET(value,flag) macro so it's a lot easier to read, it's also alightly better if you want bit > char* translation as you can use a simple char *table[] = {}; to specify names for each bit (mud coders fantasy ;-P) Rycochet -- · Phone: (+44) [0]1865 311846 *A1200, Blizzard 1260 + SCSI* · Email: rycochet@free4all.co.uk *2.6Gb hd, 33.6 Modem, 50Mb Ram* · ICQ: 13463163 *2x CD-ROM... Phillips 8833* ------------------------------------------------------------------------ PERFORM CPR ON YOUR APR! Get a NextCard Visa, in 30 seconds! Get rates as low as 0.0% Intro or 9.9% Fixed APR and no hidden fees. Apply NOW! http://click.egroups.com/1/2121/0/_/451227/_/952715832/ ------------------------------------------------------------------------ @ENDNODE @NODE "Mail_95" " casio databank watches " From: Jeffrey Barnes Date: 03-10-00 23:19:23 Subject: @{b}casio databank watches@{ub} From: Jeffrey Barnes Does any male have a casio databank watch? ____________________________________________________________________ Get your own FREE, personal Netscape WebMail account today at http://webmail.netscape.com. ------------------------------------------------------------------------ GET A NEXTCARD VISA, in 30 seconds! Get rates as low as 0.0% Intro APR and no hidden fees. Apply NOW! http://click.egroups.com/1/975/0/_/451227/_/952726854/ ------------------------------------------------------------------------ @ENDNODE @NODE "Mail_96" " Re: casio databank watches " From: Duncan Strand Date: 03-10-00 23:45:36 Subject: @{b}Re: casio databank watches@{ub} From: "Duncan Strand" > From: Jeffrey Barnes > > Does any male have a casio databank watch? Yes, why? -- Duncan Strand ICQ: 16966914 http://zebedee.tripod.com Sci-fi pictures http://www.geocities.com/aabron Lion Software If speed kills, then Windows users may live forever. ------------------------------------------------------------------------ GET A NEXTCARD VISA, in 30 seconds! Get rates as low as 2.9% Intro or 9.9% Fixed APR and no hidden fees. Apply NOW! http://click.egroups.com/1/936/0/_/451227/_/952728535/ ------------------------------------------------------------------------ @ENDNODE @NODE "Mail_97" " Re: casio databank watches " From: Stephan Huebner Date: 03-11-00 00:01:31 Subject: @{b}Re: casio databank watches@{ub} From: Stephan Huebner Hello Jeffrey On 10-Mär-00, you wrote: > Does any male have a casio databank watch? Why? Does it anything harmful to men? ;-) Greetings, Stephan -- Women: The best of this kind -- CurtGoetz ------------------------------------------------------------------------ GET A NEXTCARD VISA, in 30 seconds! Get rates as low as 2.9% Intro or 9.9% Fixed APR and no hidden fees. Apply NOW! http://click.egroups.com/1/936/0/_/451227/_/952732503/ ------------------------------------------------------------------------ @ENDNODE @NODE "Mail_98" " RE: casio databank watches " From: andrewmarkwell@ukonline.co.uk Date: 03-11-00 01:25:16 Subject: @{b}RE: casio databank watches@{ub} From: andrewmarkwell@ukonline.co.uk On Fri, 10 Mar 2000 22:45:36 -0000, "Duncan Strand" wrote: > From: "Duncan Strand" > > > From: Jeffrey Barnes > > > > Does any male have a casio databank watch? > > Yes, why? I've got a Casio Databank personal organiser which I believe is a similar technology, I've been meaning to figure out how to back up its memory on the Amiga... Andrew Markwell ------------------------------------------------------------------------ PERFORM CPR ON YOUR APR! Get a NextCard Visa, in 30 seconds! Get rates as low as 0.0% Intro or 9.9% Fixed APR and no hidden fees. Apply NOW! http://click.egroups.com/1/2121/0/_/451227/_/952734318/ ------------------------------------------------------------------------ @ENDNODE @NODE "Mail_99" " BGUI Library " From: "Alastair M. Robinson" Date: 03-11-00 16:11:29 Subject: @{b}BGUI Library@{ub} From: "Alastair M. Robinson" Hi Shane, > I've just started writing a program with the BGUI library, everything was > working fine. Though now for some reason the program won't open the library. > The strange part is I haven't actually changed my code. > I also wrote a small program just to open and close the library (test > purposes) though still no luck. > Also I've tried replacing the one in the libs directory in case it may have > been corrupt. Hasn't helped....Any suggestions? At the risk of suggesting the obvious, make sure you're not trying to open a newer version than is in Libs:. Make sure no recently installed programs have an earlier version of bgui.library in their own directory. Once one version of a library is open, requests for a newer version will fail, even if the newer version is available in libs:. The other thing to check is that all the libraries bgui needs are present and correct. I have bgui V41.8, which tries to open: dos.library, intuition.library, graphics.library, gadtools.library, utility.library, layers.library, asl.library, commodities.library, keymap.library, iffparse.library, workbench.library, locale.library datatypes.library and amigaguide.library A friend of mine had a strange problem with IBrowse recently - the window wasn't being erased properly when scrolling, so the things soon looked pretty messy! It turns out that some clever program had installed cybergraphics.library on his AGA only system! Once it was deleted, all was well again. All the best, -- Alastair M. Robinson, email:blackfive@fakenhamweb.co.uk Amiga 4000/030, 18Meg RAM, 4.3Gig HD, Mustek ScanExpress 6000SP C, Assembler (68k, PIC16C84, ATMEL AVR), PostScript, ARexx, AMOSPro(!), AmigaDOS ... Adults are just obsolete children and the hell with them. -- Theodor "Dr. Seuss" Geisel (as quoted in his obit in Time) ------------------------------------------------------------------------ GET A NEXTCARD VISA, in 30 seconds! Get rates as low as 0.0% Intro or 9.9% Fixed APR and no hidden fees. Apply NOW! http://click.egroups.com/1/937/0/_/451227/_/952816077/ ------------------------------------------------------------------------ @ENDNODE @NODE "Mail_100" " Re: What compiler.... " From: Colin Wenzel Date: 03-11-00 16:17:24 Subject: @{b}Re: What compiler....@{ub} From: Colin Wenzel Hi all, I have been using Aztec-C (K&R) for some 13 years now. It still works fine (Aztec v5.2) but it's getting a bit dated, as it doesn't have a lot of features of more modern compilers. I would like to find out what everyone is using & what is prefered these days. Also, where do you get them from... ??? ------------------------------------- Colin Wenzel. Australia. EMAIL: mailto:colstv@hotkey.net.au URL: http://www.hotkey.net.au/~colstv/ ICQ: 17608330 AMIGA: 4000T, 060/50, 150Mb RAM, O.S 3.5 --------------------------------------- ------------------------------------------------------------------------ GET A NEXTCARD VISA, in 30 seconds! Get rates as low as 2.9% Intro or 9.9% Fixed APR and no hidden fees. Apply NOW! http://click.egroups.com/1/936/0/_/451227/_/952787923/ ------------------------------------------------------------------------ @ENDNODE @NODE "Mail_101" " Re: What compiler.... " From: Stephan Huebner Date: 03-12-00 00:51:39 Subject: @{b}Re: What compiler....@{ub} From: Stephan Huebner Hello Colin On 11-Mär-00, you wrote: > It still works fine (Aztec v5.2) but it's getting a bit dated, > as it doesn't have a lot of features of more modern compilers. > > I would like to find out what everyone is using & what is > prefered these days. > > Also, where do you get them from... ??? If you don't want to spend much money you could buy the new developer CD. There's a full version of H&P StormC on it (3.0 if I remember right). Greetings, Stephan -- Gentleman: A man who is able to describe a womans' body without using his hands -- Anonymous ------------------------------------------------------------------------ PERFORM CPR ON YOUR APR! Get a NextCard Visa, in 30 seconds! Get rates as low as 0.0% Intro or 9.9% Fixed APR and no hidden fees. Apply NOW! http://click.egroups.com/1/2121/0/_/451227/_/952819308/ ------------------------------------------------------------------------ @ENDNODE @NODE "Mail_102" " Re: BGUI Library " From: Shane O'Neill Date: 03-12-00 02:11:47 Subject: @{b}Re: BGUI Library@{ub} From: Shane O'Neill Alastair Wrote: >At the risk of suggesting the obvious, make sure you're not trying to open >a newer version than is in Libs:. >Make sure no recently installed programs have an earlier version of >bgui.library in their own directory. Once one version of a library is >open, requests for a newer version will fail, even if the newer version is >available in libs:. >The other thing to check is that all the libraries bgui needs are present >and correct. I have bgui V41.8, which tries to open: >dos.library, intuition.library, graphics.library, gadtools.library, >utility.library, layers.library, asl.library, commodities.library, >keymap.library, iffparse.library, workbench.library, locale.library >datatypes.library and amigaguide.library >A friend of mine had a strange problem with IBrowse recently - the window >wasn't being erased properly when scrolling, so the things soon looked >pretty messy! It turns out that some clever program had installed >cybergraphics.library on his AGA only system! Once it was deleted, all was >well again. Hi Alastair, You may have missed my last post, I have the problem sorted. Basically I forgot to remove 41.8 from execs library list. By the way there's a memory leak in BGUI_Request from the 41.8 library. Cheers for the response. Regards, Shane. ------------------------------------------------------------------------ GET A NEXTCARD VISA, in 30 seconds! Get rates as low as 0.0% Intro or 9.9% Fixed APR and no hidden fees. Apply NOW! http://click.egroups.com/1/937/0/_/451227/_/952823593/ ------------------------------------------------------------------------ @ENDNODE @NODE "Mail_103" " Way off topic " From: Lee Atkins Date: 03-12-00 12:26:40 Subject: @{b}Way off topic@{ub} From: Lee Atkins Hello, I have a problem that has been around for about a week. I have the number of seconds passed since whenever stored as a 32 bit number. This is in an 8 bit microcontroller so I have the number stored in 4 registers like: Most significant R16 R17 R18 R19 Least significant. I have a 16 bit unsigned division routine. Is there some way someone knows of that I can do a 32 bit unsigned division on this number without writing a 32 bit division routine? Lee. ------------------------------------------------------------------------ MAXIMIZE YOUR CARD, MINIMIZE YOUR RATE! Get a NextCard Visa, in 30 seconds! Get rates as low as 0.0% Intro or 9.9% Fixed APR and no hidden fees. Apply NOW! http://click.egroups.com/1/2122/0/_/451227/_/952860636/ ------------------------------------------------------------------------ @ENDNODE @NODE "Mail_104" " Dividing ULONGs (was: Re: Way off topic) " From: Allan Odgaard Date: 03-12-00 13:36:41 Subject: @{b}Dividing ULONGs (was: Re: Way off topic)@{ub} From: Allan Odgaard On 12-Mar-00, Lee Atkins wrote: > I have a 16 bit unsigned division routine. Is there some way someone knows > of that I can do a 32 bit unsigned division on this number without writing a > 32 bit division routine? utility.library/UDivMod32 utility.library/UDivMod32 NAME UDivMod32 -- unsigned 32 by 32 bit division and modulus. (V36) SYNOPSIS quotient:remainder = UDivMod32(dividend,divisor); D0 D1 D0 D1 ULONG:ULONG UDivMod32(ULONG,ULONG); FUNCTION Divides the unsigned 32 bit dividend by the unsigned 32 bit divisor and returns an unsigned 32 bit quotient and remainder. [...] P.S. You may want to have a look at http://www.useit.com/alertbox/980906.html for info about how to write subject lines and why you should do it -- no offence! :-) ------------------------------------------------------------------------ GET A NEXTCARD VISA, in 30 seconds! Get rates as low as 0.0% Intro APR and no hidden fees. Apply NOW! http://click.egroups.com/1/975/0/_/451227/_/952869444/ ------------------------------------------------------------------------ @ENDNODE @NODE "Mail_105" " Re: What compiler.... " From: Colin Wenzel Date: 03-12-00 14:37:53 Subject: @{b}Re: What compiler....@{ub} From: Colin Wenzel On 12-Mar-00, Stephan Huebner wrote: > From: Stephan Huebner > Hello Colin > On 11-Mär-00, you wrote: > >> It still works fine (Aztec v5.2) but it's getting a bit dated, >> as it doesn't have a lot of features of more modern compilers. >> >> I would like to find out what everyone is using & what is >> prefered these days. >> >> Also, where do you get them from... ??? > > If you don't want to spend much money you could buy the new developer > CD. There's a full version of H&P StormC on it (3.0 if I remember > right). - Of course, when do you ever WANT to spend a lot of money.... ?? When is it due for release... ?? ------------------------------------- Colin Wenzel. Australia. EMAIL: mailto:colstv@hotkey.net.au URL: http://www.hotkey.net.au/~colstv/ ICQ: 17608330 AMIGA: 4000T, 060/50, 150Mb RAM, O.S 3.5 --------------------------------------- ------------------------------------------------------------------------ DON'T HATE YOUR RATE! Get a NextCard Visa, in 30 seconds! Get rates as low as 0.0% Intro or 9.9% Fixed APR and no hidden fees. Apply NOW! http://click.egroups.com/1/2120/0/_/451227/_/952868349/ ------------------------------------------------------------------------ @ENDNODE @NODE "Mail_106" " Re: What compiler.... " From: Allan Odgaard Date: 03-12-00 14:55:03 Subject: @{b}Re: What compiler....@{ub} From: Allan Odgaard On 11-Mar-00, Colin Wenzel wrote: > I would like to find out what everyone is using & what is > prefered these days. Now that's a simple question with only one answer... ;-) I think the trend of today is: StormC, SAS/C, VBCC, GCC and I've also seen people use DiceC, Maxton-C and HiSoft-C -- and I'm sure someone will blame me for not mentioning what they're using... in other words; there is no single prefered compiler system. I only have experience with StormC, but here's my subjective opinion anyhow: StormC: Nice IDE, almost uptodate compiler (i.e. C++) and average code quality with only very few bugs, despite rumours. HiSoft-C: The best IDE (of what's available on Amiga) but not very good code, and many bugs. VBCC: No IDE, only ANSI-C but very good code. GCC: Worst interface (unix) but totally uptodate with the C standard and very good code. SAS/C: Also a command line compiler, but good code and very little C++ support. How many did I manage to offend with the above statement? ;-) > Also, where do you get them from... ??? VBCC & GCC are free and the latter is on Aminet. GCC should be retrieved from http://www.ninemoons.com/ even though it may also be on Aminet. I don't think you can get SAS/C anylonger. StormC 3 is on the developer CD. Regards Allan -- URL: http://www.diku.dk/students/duff/ Your mouse has moved. WinNT must be restarted for the change to take effect. ------------------------------------------------------------------------ MAXIMIZE YOUR CARD, MINIMIZE YOUR RATE! Get a NextCard Visa, in 30 seconds! Get rates as low as 0.0% Intro or 9.9% Fixed APR and no hidden fees. Apply NOW! http://click.egroups.com/1/2122/0/_/451227/_/952869527/ ------------------------------------------------------------------------ @ENDNODE @NODE "Mail_107" " RE: What compiler.... " From: andrewmarkwell@ukonline.co.uk Date: 03-12-00 15:16:53 Subject: @{b}RE: What compiler....@{ub} From: andrewmarkwell@ukonline.co.uk On Sun, 12 Mar 2000 14:55:03 +0100, Allan Odgaard wrote: > From: Allan Odgaard > > HiSoft-C: The best IDE (of what's available on Amiga) but not very go > od code, Actually, this complier produces the smallest code out of all the ones mentioned. Andrew Markwell ------------------------------------------------------------------------ DON'T HATE YOUR RATE! Get a NextCard Visa, in 30 seconds! Get rates as low as 0.0% Intro or 9.9% Fixed APR and no hidden fees. Apply NOW! http://click.egroups.com/1/2120/0/_/451227/_/952870614/ ------------------------------------------------------------------------ @ENDNODE @NODE "Mail_108" " Re: Re: What compiler.... " From: Dirk Harlaar Date: 03-12-00 16:10:30 Subject: @{b}Re: Re: What compiler....@{ub} From: "Dirk Harlaar" Hi, [about developer CD] > Of course, when do you ever WANT to spend a lot of money.... ?? > > When is it due for release... ?? About three months ago :). Try http://www.amigainnovations.com for an Australian dealer. Last time I tried, they had it in stock. I'm pretty confident they will ship it, if you live to far away to pick it up. Dirk ------------------------------------------------------------------------ GET A NEXTCARD VISA, in 30 seconds! Get rates as low as 0.0% Intro APR and no hidden fees. Apply NOW! http://click.egroups.com/1/975/0/_/451227/_/952874093/ ------------------------------------------------------------------------ @ENDNODE @NODE "Mail_109" " StormC and Dev.- CD 2.1 Examples (printer)... " From: whose@gmx.de Date: 03-13-00 03:21:44 Subject: @{b}StormC and Dev.- CD 2.1 Examples (printer)...@{ub} From: whose@gmx.de Hello, everyone! I 've got a big problem: I started OS programming a few days before. Compiling of most examples was no problem, but the printer driver example is a bit "tricky" to me. I 've reworked the sources to compile with StromC so far, but linking the printer driver wasn't successful. Perhaps (for sure? ;) some files will not be linked to the driver executable. I tried to find out the meaning of the provided SAS make file, but I even don't know, why this thing fails in linking. It would be nice, if somebody could give me a little introduction in compiling and linking such examples (printer). Thanks in advance, Wolfgang Hosemann whose@gmx.de ------------------------------------------------------------------------ MAXIMIZE YOUR CARD, MINIMIZE YOUR RATE! Get a NextCard Visa, in 30 seconds! Get rates as low as 0.0% Intro or 9.9% Fixed APR and no hidden fees. Apply NOW! http://click.egroups.com/1/2122/0/_/451227/_/952914537/ ------------------------------------------------------------------------ @ENDNODE @NODE "Mail_110" " Re: Creating ARexx variables " From: Richard Drummond Date: 03-13-00 10:00:53 Subject: @{b}Re: Creating ARexx variables@{ub} From: "Richard Drummond" Hello Neil On 09-Mar-00, you wrote: >> This is question to do with making a C program act as an ARexx host. >> I want to be able to return the results of an ARexx command in a named >> variable, possibly a stemmed variable. How do I create create such a >> variable from C? > > AFAIK, you can only use SetRexxVar() from Amiga.lib (or something based on > the same idea/code). You can't use the result variables for it. Ah ha! I didn't notice that function tucked away inside amiga.lib. Thanks. Cheers, Rich -- Richard Drummond Staff Writer, Amiga Format mailto: richard.drummond@futurenet.co.uk pgp : http://www.drummond.u-net.com/download/richards_key.asc phone : +44 (0)1225 442244 ext 2417 ------------------------------------------------------------------------ DON'T HATE YOUR RATE! Get a NextCard Visa, in 30 seconds! Get rates as low as 0.0% Intro or 9.9% Fixed APR and no hidden fees. Apply NOW! http://click.egroups.com/1/2120/0/_/451227/_/952941888/ ------------------------------------------------------------------------ @ENDNODE @NODE "Mail_111" " Re: Way off topic " From: Duane McDonnell Date: 03-13-00 11:49:52 Subject: @{b}Re: Way off topic@{ub} From: Duane McDonnell >From: Lee Atkins >Hello, >I have a problem that has been around for about a week. >I have the number of seconds passed since whenever stored as a 32 bit >number. >This is in an 8 bit microcontroller so I have the number stored in 4 >registers like: >Most significant R16 R17 R18 R19 Least significant. >I have a 16 bit unsigned division routine. Is there some way someone knows >of that I can do a 32 bit unsigned division on this number without writing a >32 bit division routine? >Lee. > Check a book on digital circuits that explains how to design divide circuits. Once you understand the theory used to make divide circuits it should then follow how you can use an existing 16-bit divider to achieve 32-bit division (if you know how high school long division works, you already know the theory). I don't have any references handy but a search on the archives at DejaNews (www.dejanews.com) should turn up references and possibly even implementaions for various CPUs. ------------------------------------------------------------------------ GET A NEXTCARD VISA, in 30 seconds! Get rates as low as 0.0% Intro APR and no hidden fees. Apply NOW! http://click.egroups.com/1/975/0/_/451227/_/952881943/ ------------------------------------------------------------------------ @ENDNODE @NODE "Mail_112" " Re: What compiler.... " From: Matthew Fletcher Date: 03-13-00 12:01:20 Subject: @{b}Re: What compiler....@{ub} From: "Matthew Fletcher" Well here is my 2 penny`s worth,.. > I would like to find out what everyone is using & what is > prefered these days. I seem to use gcc almost all the time, probably something to do with my need to cross-compile most stuff to ppc / x86 unix. For amiga development i dont think you can get much better than Storm, if only cos its the only one being developed. (and as prev mentioned you can get it free) > AMIGA: 4000T, 060/50, 150Mb RAM, O.S 3.5 Given your system gcc will fly (it needs the mem badly), it also has a better chance of compiling anything you throw at it. Its bang upto date with ANSI C9X and AT&T C++ standards, and can produce outstanding code. -matthew ------------------------------------------------------------------------ GET A NEXTCARD VISA, in 30 seconds! Get rates as low as 2.9% Intro or 9.9% Fixed APR and no hidden fees. Apply NOW! http://click.egroups.com/1/936/0/_/451227/_/952945297/ ------------------------------------------------------------------------ @ENDNODE @NODE "Mail_113" " RE: FW: Digest Number 312 " From: roderick_maclean@agilent.com Date: 03-13-00 12:19:35 Subject: @{b}RE: FW: Digest Number 312@{ub} From: roderick_maclean@agilent.com >>> Does any male have a casio databank watch? >> >> Yes, why? > > I've got a Casio Databank personal organiser which I believe is a > similar technology, I've been meaning to figure out how to > back up its memory on the Amiga... I've got a Casio databank and because it's got no ports, no infrared and no other means of communication apart from the screen, I've found that the easiest way to 'back up' it's memory is to type everything into the computer by hand :| (and out of interest, why are you interested about males with databanks and not females?) Rod -- Rod MacLean | Roderick_MacLean@agilent.com Developer - Agilent Technologies | +44 (0131) 331-7304 ** My views do not necessarily reflect the views of Agilent! ** ------------------------------------------------------------------------ GET A NEXTCARD VISA, in 30 seconds! Get rates as low as 2.9% Intro or 9.9% Fixed APR and no hidden fees. Apply NOW! http://click.egroups.com/1/936/0/_/451227/_/952946457/ ------------------------------------------------------------------------ @ENDNODE @NODE "Mail_114" " RE: Re: What compiler.... " From: Jesper Svennevid Date: 03-13-00 13:13:21 Subject: @{b}RE: Re: What compiler....@{ub} From: Jesper Svennevid > From: Colin Wenzel > > > Hi all, > I have been using Aztec-C (K&R) for some 13 years now. > > It still works fine (Aztec v5.2) but it's getting a bit dated, > as it doesn't have a lot of features of more modern compilers. > > I would like to find out what everyone is using & what is > prefered these days. > > Also, where do you get them from... ??? There are 4 compilers that I find useful enough, and they are SAS/C, VBCC, StormC and GCC. SAS/C is getting older, and even though it got updated after termination at SAS, it still doesn't support full C++, and contains a few nasty bugs. But, it's easy to setup, and it's fast and slim. VBCC is getting better, AND it's multiplatform(big advantage), and setting it up takes almost no time at all (a few assigns and paths). Only really big flaw is that it's still a very "young" compiler, so many bugs may still lurk in the core. I've never really liked StormC, mostly because the very clumsy and slow IDE that surrounds it, and the fact that it sometimes produces really crap code. GCC is the compiler I use right now, 2.95.2 for m68k development and 2.91.66 (egcs) for PPC development. It's quite fast(the C++ compiler is getting faster aswell), and most of all, it's actively being developed by a vast range of people. The fact that it's existing on numerous platforms is a great help when you need to port your code, as everything that isn't tied to AOS(for example) compiles out of the box. SAS/C is commercial, but not sold anymore, so you need to get a copy somewhere else. VBCC is available on aminet, in a few separate packages. StormC is commerical and you can get it from Haage&Partner, either by purchasing the full package, OR by purchasing the latest DevCD(2.1). GCC is most preferably downloaded from the Geek Gadgets collection (at ftp.ninemoons.com). Be sure to fetch the 2.95.1/2 compilers in the alpha- directory, they work, even though they aren't really official. :) PPC-support is added by first installing the ppctools from the '98 release, and downloading the packs by Holger Jakob (you need the geninclude.pl from his ixemul package to convert the includes so you don't get offset-errors on ppc), and Paladin (I'm not at this moment sure of his real name :), where you get the updated compilers (cc1 and cc1plus), re-compiled libraries and a few useful includes. These includes contains errors when you compile for WarpOS, but it's quite easily fixed, just edit the warpos-part of macros.h and convert from alloca() to normal stack-allocation. -- // Jesper Svennevid : jesper@fxrealm.com Software Developer at FXRealm Studios http://www.fxrealm.com/ ------------------------------------------------------------------------ GET A NEXTCARD VISA, in 30 seconds! Get rates as low as 0.0% Intro or 9.9% Fixed APR and no hidden fees. Apply NOW! http://click.egroups.com/1/937/0/_/451227/_/952956438/ ------------------------------------------------------------------------ @ENDNODE @NODE "Mail_115" " Re: What compiler.... " From: Martyn Capewell Date: 03-13-00 13:52:36 Subject: @{b}Re: What compiler....@{ub} From: Martyn Capewell Hello Matthew On 13-Mar-00, you wrote: > From: "Matthew Fletcher" > For amiga development i dont think you can get much better than > Storm, if only cos its the only one being developed. (and as prev > mentioned you can get it free) VBCC is still developed. And is also free. Regards -- - Martyn Capewell ... but you can't expect to wield supreme executive power just because some watery tart threw a sword at you! ------------------------------------------------------------------------ GET A NEXTCARD VISA, in 30 seconds! Get rates as low as 0.0% Intro or 9.9% Fixed APR and no hidden fees. Apply NOW! http://click.egroups.com/1/937/0/_/451227/_/952951994/ ------------------------------------------------------------------------ @ENDNODE @NODE "Mail_116" " About future compatibility... " From: Gabriele Svelto Date: 03-13-00 17:27:57 Subject: @{b}About future compatibility...@{ub} From: Gabriele Svelto Hi everybody, I am using a couple of things in a proggie which I'm not sure are safe to use since they could be eliminated in new versions of AmigaOS. The first is the MEMF_CLEAR flag for the AllocVec() function, I use it when allocating (non time-critical allocations!) structures in order to get every parameter initialized to 0 (or NULL), this saves me a lot of work and avoids tons of bugs. The other is the BYTE *UserData in the Window structure, I use it for stocking data for each window the proggie opens, yes I could use an external reference for each window but I'm too lazy to do that right now. Anybody thinks that those things could be removed (if we get a new version ;-)? Thanx everybody in advance Ciao Gabriele ------------------------------------------------------------------------ MAXIMIZE YOUR CARD, MINIMIZE YOUR RATE! Get a NextCard Visa, in 30 seconds! Get rates as low as 0.0% Intro or 9.9% Fixed APR and no hidden fees. Apply NOW! http://click.egroups.com/1/2122/0/_/451227/_/953142622/ ------------------------------------------------------------------------ @ENDNODE @NODE "Mail_117" " Re: What compiler.... " From: Hynek Schlawack Date: 03-13-00 18:07:57 Subject: @{b}Re: What compiler....@{ub} From: Hynek Schlawack Hello Jesper On 13-Mär-00, you wrote: [GCC] It's quite fast(the C++ compiler is> getting faster aswell), Maybe you have a quiet fast Amiga but anyway it's a sin to call gcc fast compared to SAS/C or vbcc... Anyway, nobody seemed to mention DICE, for me the only alternative to SAS/C because it is an _Amiga_ compiler and not a cross compiler. That makes him fast and it takes not so much space. I also used gcc a while but it is deadly slow (040/40), especially the C++ part. I'm looking forward to get a SAS/C package, the best C compiler for the Amiga ever released. Yes, I also bought StormC but I think that it's really useless. Regards, -- Hynek Schlawack ++ 7Wire Technologies ++ http://me.in-berlin.de/~hys ------------------------------------------------------------------------ MAXIMIZE YOUR CARD, MINIMIZE YOUR RATE! Get a NextCard Visa, in 30 seconds! Get rates as low as 0.0% Intro or 9.9% Fixed APR and no hidden fees. Apply NOW! http://click.egroups.com/1/2122/0/_/451227/_/952970337/ ------------------------------------------------------------------------ @ENDNODE @NODE "Mail_118" " Re: Re: What compiler.... " From: Paul Hill Date: 03-13-00 21:22:00 Subject: @{b}Re: Re: What compiler....@{ub} From: "Paul Hill" On 13-Mar-00 17:07:57, Hynek Schlawack wrote: >Hello Jesper >On 13-Mär-00, you wrote: >[GCC] >It's quite fast(the C++ compiler is> getting faster aswell), >Maybe you have a quiet fast Amiga but anyway it's a sin to call gcc fast >compared to SAS/C or vbcc... VBCC is faster than GCC I agree. But I've seen VBCC take much longer (10x) than GCC with full opts enabled. It also used tons more memory. >Anyway, nobody seemed to mention DICE, for me the only alternative to SAS/C >because it is an _Amiga_ compiler and not a cross compiler. That makes him >fast and it takes not so much space. DICE is OK (I used it on my old A500 during my C studies..). The code it produces isn't always great though. Actually DICE is a generic 680x0 compiler, not strictly Amiga. >I also used gcc a while but it is deadly slow (040/40), especially the C++ >part. I'm looking forward to get a SAS/C package, the best C compiler for >the Amiga ever released. GCC still produces the best code IMHO. >Yes, I also bought StormC but I think that it's really useless. Never tried it. I tried the 1.x demo and got scared. Compilers shouldn't have windows with buttons and things :-) I couldn't work out how to dump the GUI and use a makefile. -- Paul ------------------------------------------------------------------------ GET A NEXTCARD VISA, in 30 seconds! Get rates as low as 0.0% Intro or 9.9% Fixed APR and no hidden fees. Apply NOW! http://click.egroups.com/1/937/0/_/451227/_/952981929/ ------------------------------------------------------------------------ @ENDNODE @NODE "Mail_119" " RE: Re: What compiler.... " From: Jesper Svennevid Date: 03-13-00 22:04:19 Subject: @{b}RE: Re: What compiler....@{ub} From: Jesper Svennevid > From: Hynek Schlawack > > [GCC] > > It's quite fast(the C++ compiler is> getting faster aswell), > > Maybe you have a quiet fast Amiga but anyway it's a sin to > call gcc fast compared to SAS/C or vbcc... Well, it's fast enough to me :) And as I have a ppc, I use the ppc-compiled compiler-executables which are much faster than the 68k-native ones. :) > Anyway, nobody seemed to mention DICE, for me the only > alternative to SAS/C because it is an _Amiga_ compiler and > not a cross compiler. That makes him fast and it takes not > so much space. I've never used DICE, so I couldn't have had any opinion on that perticular package. :) One nice thing with dice is that even though it is a amiga-compiler, it is compilable (the released sources, that is) on non-amiga systems, so you could develop the program on very fast systems and test it on an amiga. > I also used gcc a while but it is deadly slow (040/40), > especially the C++ > part. I'm looking forward to get a SAS/C package, the best C > compiler for > the Amiga ever released. Hm, well, C using SAS/C is good, but the C++ part leaves a lot of question-marks... no support for something as trivial as optional arguments is not that nice.. :/ > Yes, I also bought StormC but I think that it's really useless. Yeah, I've used it quite a bit when I didn't have SAS/C(way back, around StormC 1/2), and I had real nasty problems with it. StormC 3 got better, but not good enough. -- // Jesper Svennevid : jesper@fxrealm.com Software Developer at FXRealm Studios http://www.fxrealm.com/ ------------------------------------------------------------------------ MAXIMIZE YOUR CARD, MINIMIZE YOUR RATE! Get a NextCard Visa, in 30 seconds! Get rates as low as 0.0% Intro or 9.9% Fixed APR and no hidden fees. Apply NOW! http://click.egroups.com/1/2122/0/_/451227/_/952981503/ ------------------------------------------------------------------------ @ENDNODE @NODE "Mail_120" " RE: Re: What compiler.... " From: Shane O'Neill Date: 03-14-00 01:45:39 Subject: @{b}RE: Re: What compiler....@{ub} From: Shane O'Neill Hi Jesper, >From: Jesper Svennevid [snip] >I've never really liked StormC, mostly because the very clumsy and slow IDE >that surrounds it, and the fact that it sometimes produces really crap >code. Have to disagree with you here. I know a lot of people don't like the StormC IDE, though I don't mind it too much (then again most people say I'm weird ;) ) I do dislike that I can't use it in a shell though, and generally use GCC for smaller programs. The StormC IDE is very handy for larger projects. Anyway that's not what I disagree with most, StormC produce the fastest code I've seen from any Amiga compiler, sure the executables are larger than most of the other compilers, but I guess that could be down to a lot of loop unrolling. Though I only have version 2, but I'd be surprised if version 3s code took a nose dive. [snip] Regards, Shane O'Neill ------------------------------------------------------------------------ GET A NEXTCARD VISA, in 30 seconds! Get rates as low as 0.0% Intro or 9.9% Fixed APR and no hidden fees. Apply NOW! http://click.egroups.com/1/937/0/_/451227/_/952994826/ ------------------------------------------------------------------------ @ENDNODE @NODE "Mail_121" " Re: What compiler.... " From: Colin Wenzel Date: 03-14-00 02:51:25 Subject: @{b}Re: What compiler....@{ub} From: Colin Wenzel On 14-Mar-00, Paul Hill wrote: -snip- > GCC still produces the best code IMHO. > >> Yes, I also bought StormC but I think that it's really useless. > > Never tried it. I tried the 1.x demo and got scared. Compilers > shouldn't have windows with buttons and things :-) I couldn't work > out how to dump the GUI and use a makefile. Funny you mention this "MakeFile". At some time 13+ years ago, when I started using Aztec Manx C, I encountered the "Make" program, unfortunately the syntax it used made no sence to me at the time, it seemed a bizare way of process organising on an Amiga, So, I just made up a batch file to do the compile, assemble & link. It just goes in the current dir that the source code modules are in... I have never used "makefile" since. I never could see any relevant use for it, especially when it requires you to learn *another* symbolic language, I think i'm on my 10th + now, I guess my head will explode if I have to learn another one.... (:> ------------------------------------- Colin Wenzel. Australia. EMAIL: mailto:colstv@hotkey.net.au URL: http://www.hotkey.net.au/~colstv/ ICQ: 17608330 AMIGA: 4000T, 060/50, 150Mb RAM, O.S 3.5 --------------------------------------- ------------------------------------------------------------------------ GET A NEXTCARD VISA, in 30 seconds! Get rates as low as 0.0% Intro APR and no hidden fees. Apply NOW! http://click.egroups.com/1/975/0/_/451227/_/952998785/ ------------------------------------------------------------------------ @ENDNODE @NODE "Mail_122" " Better editor than se in SAS/C 6.0? " From: Neil O'Rourke Date: 03-14-00 11:09:02 Subject: @{b}Better editor than se in SAS/C 6.0?@{ub} From: "Neil O'Rourke" Hi :) I've recently bought a secondhand SAS/C 6.0, and I'm finding the supplied editor, se, woeful. It crashes a lot, corrupts text, you name it. So, having been away from the Amiga scene for several years now, I'm wondering 1) is this normal se behaviour? 2) what C editor is recommended? I use MS Visual C++ 5.0 for work and the PC side of things, and generally fine that editor a joy to use. If there's anything that comes close out there... Thanks, Neil ------------------------------------------------------------------------ MAXIMIZE YOUR CARD, MINIMIZE YOUR RATE! Get a NextCard Visa, in 30 seconds! Get rates as low as 0.0% Intro or 9.9% Fixed APR and no hidden fees. Apply NOW! http://click.egroups.com/1/2122/0/_/451227/_/953028572/ ------------------------------------------------------------------------ @ENDNODE @NODE "Mail_123" " RE: Better editor than se in SAS/C 6.0? " From: "Sealey, M." Date: 03-14-00 11:15:17 Subject: @{b}RE: Better editor than se in SAS/C 6.0?@{ub} From: "Sealey, M." > From: "Neil O'Rourke" > > Hi :) > > I've recently bought a secondhand SAS/C 6.0, and I'm finding > the supplied editor, se, woeful. It crashes a lot, corrupts text, > you name it. > > So, having been away from the Amiga scene for several years now, I'm > wondering > > 1) is this normal se behaviour? Yes ;) > 2) what C editor is recommended? I use MS Visual C++ 5.0 for > work and the PC side of things, and generally fine that editor a > joy to use. If there's anything that comes close out there... You should either go get CygnusEd or GoldED. Both are geared up as programmer's editors, and the newer GED releases have all kinds or project management tools and plugins. Dunno about CED but IIRC Olaf Barthel swears by it ;) -- Matt Sealey mws2@le.ac.uk Distributed Systems Support Computer Centre University of Leicester ------------------------------------------------------------------------ PERFORM CPR ON YOUR APR! Get a NextCard Visa, in 30 seconds! Get rates as low as 0.0% Intro or 9.9% Fixed APR and no hidden fees. Apply NOW! http://click.egroups.com/1/2121/0/_/451227/_/953029214/ ------------------------------------------------------------------------ @ENDNODE @NODE "Mail_124" " Re: What compiler.... " From: Hynek Schlawack Date: 03-14-00 12:52:58 Subject: @{b}Re: What compiler....@{ub} From: Hynek Schlawack Hello Paul On 13-Mär-00, you wrote: >> [GCC] >>> It's quite fast(the C++ compiler is> getting faster aswell), (I forgot the ">" in my last quote, sorry for it). >> Maybe you have a quiet fast Amiga but anyway it's a sin to call gcc fast >> compared to SAS/C or vbcc... > VBCC is faster than GCC I agree. But I've seen VBCC take much longer (10x) > than GCC with full opts enabled. It also used tons more memory. I never used vbcc because it seems not to be finished... >> Anyway, nobody seemed to mention DICE, for me the only alternative to >> SAS/C because it is an _Amiga_ compiler and not a cross compiler. That >> makes him fast and it takes not so much space. > DICE is OK (I used it on my old A500 during my C studies..). The code it > produces isn't always great though. That might be true. > Actually DICE is a generic 680x0 compiler, not strictly Amiga. Okay, but he is really useable on an Amiga anyway. At least developing using it makes more fun than with gcc. >> I also used gcc a while but it is deadly slow (040/40), especially the >> C++ part. I'm looking forward to get a SAS/C package, the best C compiler >> for the Amiga ever released. > GCC still produces the best code IMHO. Yes, algorithmically, but as SAS is a specific Amiga compiler, I believe that it does some system specific optimisations but anyway, the gcc code is better but waiting 30 sec for a "Hello world!" program is not acceptable. >> Yes, I also bought StormC but I think that it's really useless. > Never tried it. I tried the 1.x demo and got scared. Compilers shouldn't > have windows with buttons and things :-) I couldn't work out how to dump > the GUI and use a makefile. According to the manual it is possible but StormC is full of bugs and makes lousy code, so why use it? Let's see what StormC 4 will bring... Regards, -- Hynek Schlawack ++ 7Wire Technologies ++ http://me.in-berlin.de/~hys ------------------------------------------------------------------------ DON'T HATE YOUR RATE! Get a NextCard Visa, in 30 seconds! Get rates as low as 0.0% Intro or 9.9% Fixed APR and no hidden fees. Apply NOW! http://click.egroups.com/1/2120/0/_/451227/_/953037743/ ------------------------------------------------------------------------ @ENDNODE @NODE "Mail_125" " Re: What compiler.... " From: Hynek Schlawack Date: 03-14-00 12:55:38 Subject: @{b}Re: What compiler....@{ub} From: Hynek Schlawack Hello Jesper On 13-Mär-00, you wrote: >> I also used gcc a while but it is deadly slow (040/40), >> especially the C++ >> part. I'm looking forward to get a SAS/C package, the best C >> compiler for >> the Amiga ever released. > Hm, well, C using SAS/C is good, but the C++ part leaves a lot of > question-marks... no support for something as trivial as optional > arguments is not that nice.. :/ I suppose this mailinglist is called amiga-c and not amiga-c++ ;-)? I agree that gcc is the opnly useable C++ compiler for the Amiga, but he is simply to slow. >> Yes, I also bought StormC but I think that it's really useless. > Yeah, I've used it quite a bit when I didn't have SAS/C(way back, around > StormC 1/2), and I had real nasty problems with it. StormC 3 got better, > but not good enough. There are just too much bugs in it, eg. "long long" handling etc., I had really big problems with it. Regards, -- Hynek Schlawack ++ 7Wire Technologies ++ http://me.in-berlin.de/~hys ------------------------------------------------------------------------ MAXIMIZE YOUR CARD, MINIMIZE YOUR RATE! Get a NextCard Visa, in 30 seconds! Get rates as low as 0.0% Intro or 9.9% Fixed APR and no hidden fees. Apply NOW! http://click.egroups.com/1/2122/0/_/451227/_/953037744/ ------------------------------------------------------------------------ @ENDNODE @NODE "Mail_126" " Re: Better editor than se in SAS/C 6.0? " From: Hynek Schlawack Date: 03-14-00 13:11:32 Subject: @{b}Re: Better editor than se in SAS/C 6.0?@{ub} From: Hynek Schlawack Hello M. On 14-Mär-00, you wrote: >> 2) what C editor is recommended? I use MS Visual C++ 5.0 for >> work and the PC side of things, and generally fine that editor a >> joy to use. If there's anything that comes close out there... > You should either go get CygnusEd or GoldED. Both are geared up > as programmer's editors, and the newer GED releases have all kinds > or project management tools and plugins. Dunno about CED but > IIRC Olaf Barthel swears by it ;) He even develops it :-). Regards, -- Hynek Schlawack ++ 7Wire Technologies ++ http://me.in-berlin.de/~hys ------------------------------------------------------------------------ GET A NEXTCARD VISA, in 30 seconds! Get rates as low as 2.9% Intro or 9.9% Fixed APR and no hidden fees. Apply NOW! http://click.egroups.com/1/936/0/_/451227/_/953037750/ ------------------------------------------------------------------------ @ENDNODE @NODE "Mail_127" " Re: Re: What compiler.... " From: Paul Hill Date: 03-14-00 13:54:44 Subject: @{b}Re: Re: What compiler....@{ub} From: Paul Hill Colin Wenzel wrote: > On 14-Mar-00, Paul Hill wrote: > -snip- > > GCC still produces the best code IMHO. > > > >> Yes, I also bought StormC but I think that it's really useless. > > > > Never tried it. I tried the 1.x demo and got scared. Compilers > > shouldn't have windows with buttons and things :-) I couldn't work > > out how to dump the GUI and use a makefile. > > Funny you mention this "MakeFile". > > At some time 13+ years ago, when I started using Aztec Manx C, > I encountered the "Make" program, unfortunately the syntax it used > made no sence to me at the time, it seemed a bizare way of > process organising on an Amiga, So, I just made up a batch file > to do the compile, assemble & link. > > It just goes in the current dir that the source code modules are in... But the method you're using is really slow as it will have to compile *every* source file every time. A makefile will only compile the programs that have changed. You also tell it about dependencies so if you change a header file it knows what program files to change. > I have never used "makefile" since. You must have plenty of time :-) > I never could see any relevant use for it, especially when > it requires you to learn *another* symbolic language, > I think i'm on my 10th + now, > I guess my head will explode if I have to learn another one.... > (:> Tried REBOL yet :-) Just use a makefile-generator tool. There's several around that scan your source code and generate a makefile with dependencies. -- Paul ------------------------------------------------------------------------ PERFORM CPR ON YOUR APR! Get a NextCard Visa, in 30 seconds! Get rates as low as 0.0% Intro or 9.9% Fixed APR and no hidden fees. Apply NOW! http://click.egroups.com/1/2121/0/_/451227/_/953041166/ ------------------------------------------------------------------------ @ENDNODE @NODE "Mail_128" " Re: Re: What compiler.... " From: Paul Hill Date: 03-14-00 14:00:33 Subject: @{b}Re: Re: What compiler....@{ub} From: Paul Hill Hynek Schlawack wrote: > Hello Paul > On 13-Mär-00, you wrote: > > >> Maybe you have a quiet fast Amiga but anyway it's a sin to call gcc fast > >> compared to SAS/C or vbcc... > > VBCC is faster than GCC I agree. But I've seen VBCC take much longer (10x) > > than GCC with full opts enabled. It also used tons more memory. > > I never used vbcc because it seems not to be finished... I've only started using it recently. It seems pretty finished to me. I managed to compile a large project of mine without too many problems. > >> I also used gcc a while but it is deadly slow (040/40), especially the > >> C++ part. I'm looking forward to get a SAS/C package, the best C compiler > >> for the Amiga ever released. > > GCC still produces the best code IMHO. > > Yes, algorithmically, but as SAS is a specific Amiga compiler, I believe > that it does some system specific optimisations but anyway, the gcc code is > better but waiting 30 sec for a "Hello world!" program is not acceptable. More like 3 seconds... The trick with GCC is to make sure you have a large disk cache (or caching filesystem). I don't see why an 'Amiga compiler' would produce better code than a generic 680x0 compiler. Having said that it seems like the 680x0 GCC is lacking behind the other targets due to lack of developers. -- Paul ------------------------------------------------------------------------ DON'T HATE YOUR RATE! Get a NextCard Visa, in 30 seconds! Get rates as low as 0.0% Intro or 9.9% Fixed APR and no hidden fees. Apply NOW! http://click.egroups.com/1/2120/0/_/451227/_/953041168/ ------------------------------------------------------------------------ @ENDNODE @NODE "Mail_129" " Re: Re: What compiler.... " From: "Andrzej J. Debicki" Date: 03-14-00 14:02:42 Subject: @{b}Re: Re: What compiler....@{ub} From: "Andrzej J. Debicki" Hello Hynek, Tuesday, March 14, 2000, you wrote: > I never used vbcc because it seems not to be finished... Tell me the definition of "finished compiler" or "finished program". I suppose, evry soft I tell you will not be finished regarding the definition :-))). I use vbcc and I think that even in the stage of "not finished" it is quite great prog. -- _ # Andrzej J. Debicki # A12ooT 66MB \\ _ # PANS - Polish Amiga News Squad # 6o3e/o6o \\// # denbi@power-amiga.prv.pl # BVisionPPC \/ # http://www.power-amiga.prv.pl # AmigaOS 3.5 ----------------------------------------------------------- ------------------------------------------------------------------------ MAXIMIZE YOUR CARD, MINIMIZE YOUR RATE! Get a NextCard Visa, in 30 seconds! Get rates as low as 0.0% Intro or 9.9% Fixed APR and no hidden fees. Apply NOW! http://click.egroups.com/1/2122/0/_/451227/_/953038991/ ------------------------------------------------------------------------ @ENDNODE @NODE "Mail_130" " Re: What compiler.... " From: Hynek Schlawack Date: 03-14-00 14:27:59 Subject: @{b}Re: What compiler....@{ub} From: Hynek Schlawack Hello Andrzej On 14-Mär-00, you wrote: >> I never used vbcc because it seems not to be finished... > Tell me the definition of "finished compiler" or "finished program". I > suppose, evry soft I tell you will not be finished regarding the > definition :-))). > I use vbcc and I think that even in the stage of "not finished" it is > quite great prog. Sorry but in the docs I always read, that the current version isn't finished pp...I'm giving vbcc a chance right now, but I don't believe that he will bring me away from DICE. Regards, -- Hynek Schlawack ++ 7Wire Technologies ++ http://me.in-berlin.de/~hys ------------------------------------------------------------------------ MAXIMIZE YOUR CARD, MINIMIZE YOUR RATE! Get a NextCard Visa, in 30 seconds! Get rates as low as 0.0% Intro or 9.9% Fixed APR and no hidden fees. Apply NOW! http://click.egroups.com/1/2122/0/_/451227/_/953049005/ ------------------------------------------------------------------------ @ENDNODE @NODE "Mail_131" " Re: Better editor than se in SAS/C 6.0? " From: Matthew Fletcher Date: 03-14-00 14:28:42 Subject: @{b}Re: Better editor than se in SAS/C 6.0?@{ub} From: "Matthew Fletcher" > > You should either go get CygnusEd or GoldED. Both are geared up > > as programmer's editors, and the newer GED releases have all kinds > > or project management tools and plugins. Dunno about CED but > > IIRC Olaf Barthel swears by it ;) > > He even develops it :-). A made-up version of golded (4.8) is on the latest aminetCD,. and i think a demo of ced was on one a while ago. I just use golded cos its the only editor that will pop a list of function names. -matthew ------------------------------------------------------------------------ GET A NEXTCARD VISA, in 30 seconds! Get rates as low as 0.0% Intro or 9.9% Fixed APR and no hidden fees. Apply NOW! http://click.egroups.com/1/937/0/_/451227/_/953040528/ ------------------------------------------------------------------------ @ENDNODE @NODE "Mail_132" " Re: What compiler.... " From: Allan Odgaard Date: 03-14-00 14:29:14 Subject: @{b}Re: What compiler....@{ub} From: Allan Odgaard On 13-Mar-00, Paul Hill wrote: >> Yes, I also bought StormC but I think that it's really useless. Useless how? Except for GCC then it's the most up-to-date compiler. If you want to program C++ then all compilers except StormC and GCC are useless in the objective sense of the word! > Never tried it. I tried the 1.x demo and got scared. Compilers shouldn't > have windows with buttons and things :-) StormC (the executable) has no GUI, it takes a source file and outputs the compiled object file. Similar for StormAsm, StormLink etc. You probably started StormShell, which is, surprise: the graphical shell. > I couldn't work out how to dump the GUI and use a makefile. That's ironical. You want to use the most primitive interface, which by definition should only be used by experts, and then you complain that you weren't able to do this ;-) Personally I've never been motivated to program C because I knew that before I could do so I'd first need to learn how to setup a compiler system with makefiles, includes etc. -- ~3½ year ago a friend of mine gave me the StormC-Demo, and I was quite amazed that I could simply press 'new', write the famous "hello world"-line and then press 'run'. If it hadn't been for that then I don't know what my programming status would be today... so I'm quite grateful for the existence of StormC :-) Regards Allan -- URL: http://www.diku.dk/students/duff/ I've never forgot my 1st Windows installation. But I'm still trying... ------------------------------------------------------------------------ DON'T HATE YOUR RATE! Get a NextCard Visa, in 30 seconds! Get rates as low as 0.0% Intro or 9.9% Fixed APR and no hidden fees. Apply NOW! http://click.egroups.com/1/2120/0/_/451227/_/953040647/ ------------------------------------------------------------------------ @ENDNODE @NODE "Mail_133" " RE: Re: What compiler.... " From: Matthew Fletcher Date: 03-14-00 14:36:28 Subject: @{b}RE: Re: What compiler....@{ub} From: "Matthew Fletcher" > > Anyway, nobody seemed to mention DICE, for me the only > > alternative to SAS/C because it is an _Amiga_ compiler and > > not a cross compiler. That makes him fast and it takes not > > so much space. Well i use it for short programs, cos is blindingly fast. But although i feel ill every time i go near iostream, c++ (even in amiga-c) is a major ommision. The chances are vbcc, sas/s, dice, northc (anyone mention that ?), or indeed Hisoft/Maxon, etc are just not upto modern c/++ standards and possably never will be. Onother question is how exactly will any of the above compilers cope with Elate OO code ?, chances are they wont. -matthew ------------------------------------------------------------------------ DON'T HATE YOUR RATE! Get a NextCard Visa, in 30 seconds! Get rates as low as 0.0% Intro or 9.9% Fixed APR and no hidden fees. Apply NOW! http://click.egroups.com/1/2120/0/_/451227/_/953040998/ ------------------------------------------------------------------------ @ENDNODE @NODE "Mail_134" " RE: Re: Better editor than se in SAS/C 6.0? " From: "Sealey, M." Date: 03-14-00 14:42:41 Subject: @{b}RE: Re: Better editor than se in SAS/C 6.0?@{ub} From: "Sealey, M." > A made-up version of golded (4.8) is on the latest aminetCD,. and i > think a demo of ced was on one a while ago. > > I just use golded cos its the only editor that will pop a list of > function names. Which reminds me, anyone know of a place where I can buy GoldED 6? I'll need it to continue my Amiga C programming (scraping back on topic) but nowhere around here sells it. There are some suppliers listed at http://golded.members.tripod.com but I haven't got a clue how good they might be.. -- Matt Sealey mws2@le.ac.uk Distributed Systems Support Computer Centre University of Leicester ------------------------------------------------------------------------ DON'T HATE YOUR RATE! Get a NextCard Visa, in 30 seconds! Get rates as low as 0.0% Intro or 9.9% Fixed APR and no hidden fees. Apply NOW! http://click.egroups.com/1/2120/0/_/451227/_/953041412/ ------------------------------------------------------------------------ @ENDNODE @NODE "Mail_135" " Re: Re: What compiler.... " From: Andrew Ballard Date: 03-14-00 14:45:34 Subject: @{b}Re: Re: What compiler....@{ub} From: Andrew Ballard » »Funny you mention this "MakeFile". » »At some time 13+ years ago, when I started using Aztec Manx C, »I encountered the "Make" program, unfortunately the syntax it used »made no sence to me at the time, it seemed a bizare way of »process organising on an Amiga, So, I just made up a batch file »to do the compile, assemble & link. » »It just goes in the current dir that the source code modules are in... » »I have never used "makefile" since. » »I never could see any relevant use for it, especially when »it requires you to learn *another* symbolic language, »I think i'm on my 10th + now, »I guess my head will explode if I have to learn another one.... » (:> » Makefiles are useful for larger projects where you have programs made from many files. It removes unnecessary compiling if you just alter one .c file. They are simply a list of rules of how to produce each file, and what files it is made from. If it knows what files it is made from, it can recursively work back and make those files using their rules. I havn't seen Storm C properly but I must say I would prefer some sort of GUI makefile system over command line based make. Andy. ------------------------------------------------------------------------ MAXIMIZE YOUR CARD, MINIMIZE YOUR RATE! Get a NextCard Visa, in 30 seconds! Get rates as low as 0.0% Intro or 9.9% Fixed APR and no hidden fees. Apply NOW! http://click.egroups.com/1/2122/0/_/451227/_/953041597/ ------------------------------------------------------------------------ @ENDNODE @NODE "Mail_136" " Re: Re: What compiler.... " From: "Francis G. Loch" Date: 03-14-00 14:54:29 Subject: @{b}Re: Re: What compiler....@{ub} From: "=?iso-8859-1?q?Francis=20G.=20Loch?=" Hi there, --- Hynek Schlawack wrote: > Hello Paul > > [SNIP] > > According to the manual it is possible but StormC is > full of bugs and makes > lousy code, so why use it? Let's see what StormC 4 > will bring... > I'd love to see what StormC 4 is like, but it's too expensive for my budget. Maybe it will be given away for free on the next version of the Developer's CD (here's hoping! ;). Kind regards, Francis. ===== e-mail:- fgloch@yahoo.com web-site:- http://www.geocities.com/fgloch __________________________________________________ Do You Yahoo!? Talk to your friends online with Yahoo! Messenger. http://im.yahoo.com ------------------------------------------------------------------------ MAXIMIZE YOUR CARD, MINIMIZE YOUR RATE! Get a NextCard Visa, in 30 seconds! Get rates as low as 0.0% Intro or 9.9% Fixed APR and no hidden fees. Apply NOW! http://click.egroups.com/1/2122/0/_/451227/_/953042070/ ------------------------------------------------------------------------ @ENDNODE @NODE "Mail_137" " Re: What compiler.... " From: Hynek Schlawack Date: 03-14-00 17:00:02 Subject: @{b}Re: What compiler....@{ub} From: Hynek Schlawack Am 14-Mär-00 schrieb Allan Odgaard: >>> Yes, I also bought StormC but I think that it's really useless. > Useless how? Except for GCC then it's the most up-to-date compiler. If you > want to program C++ then all compilers except StormC and GCC are useless > in the objective sense of the word! Yes, but I want to program C! And StormC doesn't support STL...at least not in the current version. >> Never tried it. I tried the 1.x demo and got scared. Compilers shouldn't >> have windows with buttons and things :-) > StormC (the executable) has no GUI, it takes a source file and outputs the > compiled object file. Similar for StormAsm, StormLink etc. You probably > started StormShell, which is, surprise: the graphical shell. Which is not very powerful and makes often exhibitions to India... >> I couldn't work out how to dump the GUI and use a makefile. > That's ironical. You want to use the most primitive interface, which by > definition should only be used by experts, and then you complain that you > weren't able to do this ;-) *g*. I suppose he has no printed manual. > Personally I've never been motivated to program C because I knew that > before I could do so I'd first need to learn how to setup a compiler > system with makefiles, includes etc. -- ~3½ year ago a friend of mine gave > me the StormC-Demo, and I was quite amazed that I could simply press > 'new', write the famous "hello world"-line and then press 'run'. If it > hadn't been for that then I don't know what my programming status would be > today... so I'm quite grateful for the existence of StormC :-) Hm, that might be true, but after a while using the Storm IDE I needed some functions that it couldn't offer to me and so I went first to gcc, then to DICE and today I recompiled my current project with vbcc :-). Gruss, -- Hynek Schlawack ++ 7Wire Technologies ++ http://me.in-berlin.de/~hys ------------------------------------------------------------------------ PERFORM CPR ON YOUR APR! Get a NextCard Visa, in 30 seconds! Get rates as low as 0.0% Intro or 9.9% Fixed APR and no hidden fees. Apply NOW! http://click.egroups.com/1/2121/0/_/451227/_/953053126/ ------------------------------------------------------------------------ @ENDNODE @NODE "Mail_138" " Re: What compiler.... " From: Allan Odgaard Date: 03-14-00 18:16:37 Subject: @{b}Re: What compiler....@{ub} From: Allan Odgaard On 14-Mar-00, Andrew Ballard wrote: > I havn't seen Storm C properly but I must say I would prefer some sort of GUI > makefile system over command line based make. StormC will watch all dependencies for you. It also offers a GUI for all compiler, assembler and linker settings, including such things as "link library / executable", which I don't think is trivial with a makefile. Regards Allan ------------------------------------------------------------------------ DON'T HATE YOUR RATE! Get a NextCard Visa, in 30 seconds! Get rates as low as 0.0% Intro or 9.9% Fixed APR and no hidden fees. Apply NOW! http://click.egroups.com/1/2120/0/_/451227/_/953054388/ ------------------------------------------------------------------------ @ENDNODE @NODE "Mail_139" " Re: What compiler.... " From: Allan Odgaard Date: 03-14-00 18:20:22 Subject: @{b}Re: What compiler....@{ub} From: Allan Odgaard On 14-Mar-00, Hynek Schlawack wrote: Let me start by apologizing for spending so much bandwidth on defending StormC. I'm just tied of seeing the most modern and only commercial compiler on the Amiga being accused of being crappy and buggy from people who have never really tried it... [ StormC ] > There are just too much bugs in it, eg. "long long" handling etc., I had > really big problems with it. Which problems? I just tried it and here it seems to work just fine with simple (64 bit) arithmetic. Did you report the problems? H&P has a mailinglist where one should report bugs, because they can only fix the things they are aware off. It's really rare that I see people report bugs on this mailinglist, but I constantly see people accuse it of being buggy and producing crappy code, even though I really doubt people have actually looked at the code, they're just passing on what they heard from others. Regarding the effectiveness of the produced code then this is getting better, have in mind that StormC is a relatively young product, and H&P's efforts are much better spent implementing full ISO/ANSI-C++, STL, posix or whatever it's all called -- the quality of the code comes second, and in the end it has much more to do with *your* code and algorithms than with what the compiler outputs, I think I've proven that with the few speed-critical things which are on my homepage, such as HTMLview, the LayerDemo, ImageManager.library etc. which are all compiled with StormC. Also, I don't even think SAS/C supports 64 bit arithmetic, so if your problems with this were the only problems, then I see no need to drop StormC, as falling back on SAS/C won't really make the situation any better. Regards Allan ------------------------------------------------------------------------ GET A NEXTCARD VISA, in 30 seconds! Get rates as low as 2.9% Intro or 9.9% Fixed APR and no hidden fees. Apply NOW! http://click.egroups.com/1/936/0/_/451227/_/953054467/ ------------------------------------------------------------------------ @ENDNODE @NODE "Mail_140" " Re: What compiler.... " From: Hynek Schlawack Date: 03-14-00 18:37:19 Subject: @{b}Re: What compiler....@{ub} From: Hynek Schlawack Am 14-Mär-00 schrieb Allan Odgaard: > Let me start by apologizing for spending so much bandwidth on defending > StormC. I'm just tied of seeing the most modern and only commercial > compiler on the Amiga being accused of being crappy and buggy from people > who have never really tried it... I developed using StormC 3.0 for aprox. one year. I think *that* is enough. > [ StormC ] >> There are just too much bugs in it, eg. "long long" handling etc., I had >> really big problems with it. > Which problems? I just tried it and here it seems to work just fine with > simple (64 bit) arithmetic. Sorry, but I had to use a certain number of 64 bit longs and it worked not! I had and have to write my programs for a BBS system I'm also developing on and I made enough experiences to call StormC useless. The problems were that certain operations made the vars become zero...nice eh? That were time functions for the bbs's kernel and I had to use long long. I heard from our project leader that it has something to do with registers. > Did you report the problems? H&P has a mailinglist where one should report Yes I did. > bugs, because they can only fix the things they are aware off. It's really > rare that I see people report bugs on this mailinglist, but I constantly I reported it directly to Mr. Haage. > see people accuse it of being buggy and producing crappy code, even though > I really doubt people have actually looked at the code, they're just > passing on what they heard from others. Regarding the effectiveness of the A friend of mine has reported more than enough bugs and the only answer he got was that they will it fix asap...but they didn't. Bert Schulzki is his name if you remember him. He disassembled the crapy code StormC has produced and has pointed out directly the errors StormC made. But when was the last update? And do for example the precompiled headers work? No, they just crash and Jochen Becher told me they know it...so why didn't they fix it? Or the many errors in the manual...Please don't offend me just because I'm pointing something out about your favorit compiler. I have also paid very much money for it. > produced code then this is getting better, have in mind that StormC is a > relatively young product, and H&P's efforts are much better spent > implementing full ISO/ANSI-C++, STL, posix or whatever it's all called -- Where is STL??? > the quality of the code comes second, and in the end it has much more to > do with *your* code and algorithms than with what the compiler outputs, I I wouldn't say the code is bad beacause I cannot prove it, but I have fought too much nights with its bugs to use it anymore. The quality of the code isn't my main creteria, but if the compiler produces wrong code it actually *is* a problem! > think I've proven that with the few speed-critical things which are on my > homepage, such as HTMLview, the LayerDemo, ImageManager.library etc. which > are all compiled with StormC. I haven't critized the code but everything else. > Also, I don't even think SAS/C supports 64 bit arithmetic, so if your 64 bit isn't the only problem. The shell of StormC is crashing all the time a.s.o. > problems with this were the only problems, then I see no need to drop > StormC, as falling back on SAS/C won't really make the situation any > better. As I stated I'm waiting for StormC 4, but if it will be like all other H&P products, I have no hope. Gruss, -- Hynek Schlawack ++ 7Wire Technologies ++ http://me.in-berlin.de/~hys ------------------------------------------------------------------------ GET A NEXTCARD VISA, in 30 seconds! Get rates as low as 0.0% Intro or 9.9% Fixed APR and no hidden fees. Apply NOW! http://click.egroups.com/1/937/0/_/451227/_/953055508/ ------------------------------------------------------------------------ @ENDNODE @NODE "Mail_141" " Re: Re: What compiler.... " From: Bart King Date: 03-14-00 21:42:12 Subject: @{b}Re: Re: What compiler....@{ub} From: Bart King Colin Wenzel wrote: > I would like to find out what everyone is using & what is > prefered these days. I use StormC. Despite what everyone else seems to have said about StormC it gives me no trouble at all - even though I am compiling about 2.5 MB of source in about 70 code modules (and it largely re-compiles under Windows, sans Amiga-specific stuff). I also have SAS/C, but I'm a sucker for user interfaces. Plus it seems a hell of a lot slower than StormC at both compiling and in generated code (takes about 35 minutes to do a full compile in SAS, about 7 in StormC on my A4000/040 (33mhz), 112 MB). Another reason I use Storm is for WarpOS support... er, even though I don't have a PPC card, but people I know do :) VBCC and GCC I've used in the past without fault, although I just can't stand makefiles, such a pain in the ass to keep the dependances up-to-date when you have about 50 header files... -- Bart King Developer http://www.bartman.demon.co.uk (not up yet - soon) ------------------------------------------------------------------------ PERFORM CPR ON YOUR APR! Get a NextCard Visa, in 30 seconds! Get rates as low as 0.0% Intro or 9.9% Fixed APR and no hidden fees. Apply NOW! http://click.egroups.com/1/2121/0/_/451227/_/953066884/ ------------------------------------------------------------------------ @ENDNODE @NODE "Mail_142" " Re: Better editor than se in SAS/C 6.0? " From: Colin Wenzel Date: 03-15-00 02:23:32 Subject: @{b}Re: Better editor than se in SAS/C 6.0?@{ub} From: Colin Wenzel On 14-Mar-00, Matthew Fletcher wrote: > From: "Matthew Fletcher" >> > You should either go get CygnusEd or GoldED. Both are geared up >> > as programmer's editors, and the newer GED releases have all kinds >> > or project management tools and plugins. Dunno about CED but >> > IIRC Olaf Barthel swears by it ;) >> >> He even develops it :-). > > A made-up version of golded (4.8) is on the latest aminetCD,. and i > think a demo of ced was on one a while ago. > > I just use golded cos its the only editor that will pop a list of > function names. > > -matthew I still use TxEd if anyone is interested, it works for me !!..... ------------------------------------- Colin Wenzel. Australia. EMAIL: mailto:colstv@hotkey.net.au URL: http://www.hotkey.net.au/~colstv/ ICQ: 17608330 AMIGA: 4000T, 060/50, 150Mb RAM, O.S 3.5 --------------------------------------- ------------------------------------------------------------------------ GET A NEXTCARD VISA, in 30 seconds! Get rates as low as 0.0% Intro or 9.9% Fixed APR and no hidden fees. Apply NOW! http://click.egroups.com/1/937/0/_/451227/_/953084695/ ------------------------------------------------------------------------ @ENDNODE @NODE "Mail_143" " RE: Re: What compiler.... " From: Volker Barthelmann Date: 03-15-00 02:42:01 Subject: @{b}RE: Re: What compiler....@{ub} From: Volker Barthelmann On Tue, 14 Mar 2000, Shane O'Neill wrote: > Anyway that's not what I disagree with most, StormC produce the fastest code I've > seen from any Amiga compiler, sure the executables are larger than most of the other > compilers, but I guess that could be down to a lot of loop unrolling. > Though I only have version 2, but I'd be surprised if version 3s code took a nose dive. I think it's safe to say that if your code is fastest with StormC that's definitely not the rule. Also, gcc and vbcc are the only Amiga compilers I have seen doing loop-unrolling. I have not seen StormC (up to 3.0) doing any loop optimizations like loop-unrolling, strength-reduction, invariant-code-motion or induction-variable-elimination. I'd be curious to see an example where StormC generates faster code than all other compilers. Volker ------------------------------------------------------------------------ GET A NEXTCARD VISA, in 30 seconds! Get rates as low as 2.9% Intro or 9.9% Fixed APR and no hidden fees. Apply NOW! http://click.egroups.com/1/936/0/_/451227/_/953086588/ ------------------------------------------------------------------------ @ENDNODE @NODE "Mail_144" " Re: What compiler / MakeFile " From: Colin Wenzel Date: 03-15-00 02:42:05 Subject: @{b}Re: What compiler / MakeFile@{ub} From: Colin Wenzel On 14-Mar-00, Paul Hill wrote: -snip- >> Funny you mention this "MakeFile". >> >> At some time 13+ years ago, when I started using Aztec Manx C, >> I encountered the "Make" program, unfortunately the syntax it used >> made no sence to me at the time, it seemed a bizare way of >> process organising on an Amiga, So, I just made up a batch file >> to do the compile, assemble & link. >> >> It just goes in the current dir that the source code modules are >> in... > > But the method you're using is really slow as it will have to compile > *every* source file every time. A makefile will only compile the > programs that have changed. You also tell it about dependencies so if > you change a header file it knows what program files to change. > >> I have never used "makefile" since. > > You must have plenty of time :-) > Enclosed is a program I use to check the archive bit to see if any of the modules have changed... (Incl source...) When I compile them to an object file, I protect the source file with +A ------------------------------------- Colin Wenzel. Australia. EMAIL: mailto:colstv@hotkey.net.au URL: http://www.hotkey.net.au/~colstv/ ICQ: 17608330 AMIGA: 4000T, 060/50, 150Mb RAM, O.S 3.5 --------------------------------------- ------------------------------------------------------------------------ DON'T HATE YOUR RATE! Get a NextCard Visa, in 30 seconds! Get rates as low as 0.0% Intro or 9.9% Fixed APR and no hidden fees. Apply NOW! http://click.egroups.com/1/2120/0/_/451227/_/953084694/ ------------------------------------------------------------------------ @ENDNODE @NODE "Mail_145" " Re: Re: What compiler.... " From: Volker Barthelmann Date: 03-15-00 03:10:42 Subject: @{b}Re: Re: What compiler....@{ub} From: Volker Barthelmann On Tue, 14 Mar 2000, Hynek Schlawack wrote: > Sorry but in the docs I always read, that the current version isn't finished > pp...I'm giving vbcc a chance right now, but I don't believe that he will > bring me away from DICE. Well..I think the documentation of vbcc 0.7 only says that the optimizer is not finished (something DICE doesn't have at all). There's even an item about the version number in the FAQ. :-) Indeed, I do not consider vbcc finished yet, because it does not yet have all the features I'd like it to have (the other compilers don't have these features either). However, this does not mean it's less complete or usable - many people seem to use it quite succesfully. But I have, for example, pretty precise targets regarding optimizations. vbcc 0.7 only implements part of the optimizations which had been planned - e.g. on a CyberstormPPC vbcc 0.7 gets about 400000 dhrystones/s whereas my current version gets about 730000 (measured by Frank Wille). So, the "finishedness" relates more to my own private goals rather than completeness or stability. Volker ------------------------------------------------------------------------ GET A NEXTCARD VISA, in 30 seconds! Get rates as low as 0.0% Intro or 9.9% Fixed APR and no hidden fees. Apply NOW! http://click.egroups.com/1/937/0/_/451227/_/953086585/ ------------------------------------------------------------------------ @ENDNODE @NODE "Mail_146" " Re: Re: What compiler.... " From: Laura Vance Date: 03-15-00 05:10:15 Subject: @{b}Re: Re: What compiler....@{ub} From: Laura Vance Ok, this is getting old. Every time that this same argument comes up, no matter if it's in a mailing list, in a usenet newsgroup, or anywhere else, it's always the same. Nobody is going to change anyone's mind about any compiler, and odds are that whoever asked the original question just wanted the list of compilers to make their own decision. Compiler wars are about as interesting as discussions about religion or politics, we all have our own favorite and there is nothing that anyone can say to change anyone else's mind. Sorry, but I get tired of the same old arguments every time this discussion comes up. later, Laura On Tue, 14 Mar 2000, Bart King wrote about Re: [amiga-c] Re: What compiler.... > From: Bart King > > Colin Wenzel wrote: > > > I would like to find out what everyone is using & what is > > prefered these days. > > I use StormC. Despite what everyone else seems to have said about > StormC it gives me no trouble at all - even though I am compiling about > 2.5 MB of source in about 70 code modules (and it largely re-compiles > under Windows, sans Amiga-specific stuff). > > I also have SAS/C, but I'm a sucker for user interfaces. Plus it seems > a hell of a lot slower than StormC at both compiling and in generated > code (takes about 35 minutes to do a full compile in SAS, about 7 in > StormC on my A4000/040 (33mhz), 112 MB). > > Another reason I use Storm is for WarpOS support... er, even though I > don't have a PPC card, but people I know do :) > > VBCC and GCC I've used in the past without fault, although I just can't > stand makefiles, such a pain in the ass to keep the dependances > up-to-date when you have about 50 header files... > > -- > Bart King > Developer > http://www.bartman.demon.co.uk (not up yet - soon) > > > ------------------------------------------------------------------------ > PERFORM CPR ON YOUR APR! > Get a NextCard Visa, in 30 seconds! Get rates as low as > 0.0% Intro or 9.9% Fixed APR and no hidden fees. > Apply NOW! > http://click.egroups.com/1/2121/0/_/451227/_/953066884/ > ------------------------------------------------------------------------ -- \|/ (o O) ---oOO(_)OOo--- Peekaboo!!! ------------------------------------------------------------------------ GET A NEXTCARD VISA, in 30 seconds! Get rates as low as 2.9% Intro or 9.9% Fixed APR and no hidden fees. Apply NOW! http://click.egroups.com/1/936/0/_/451227/_/953094071/ ------------------------------------------------------------------------ @ENDNODE @NODE "Mail_147" " RE: Re: What compiler.... " From: BOYER Laurent Date: 03-15-00 08:34:45 Subject: @{b}RE: Re: What compiler....@{ub} From: Laurent Boyer Yes, you are right, if they can't tell which is the best compiler, then they must make a new one and not borrow everybody ... From: Laura Vance Ok, this is getting old. Every time that this same argument comes up, no matter if it's in a mailing list, in a usenet newsgroup, or anywhere else, it's always the same. Nobody is going to change anyone's mind about any compiler, and odds are that whoever asked the original question just wanted the list of compilers to make their own decision. Compiler wars are about as interesting as discussions about religion or politics, we all have our own favorite and there is nothing that anyone can say to change anyone else's mind. Sorry, but I get tired of the same old arguments every time this discussion comes up. later, Laura On Tue, 14 Mar 2000, Bart King wrote about Re: [amiga-c] Re: What compiler.... > From: Bart King > > Colin Wenzel wrote: > > > I would like to find out what everyone is using & what is > > prefered these days. > > I use StormC. Despite what everyone else seems to have said about > StormC it gives me no trouble at all - even though I am compiling about > 2.5 MB of source in about 70 code modules (and it largely re-compiles > under Windows, sans Amiga-specific stuff). > > I also have SAS/C, but I'm a sucker for user interfaces. Plus it seems > a hell of a lot slower than StormC at both compiling and in generated > code (takes about 35 minutes to do a full compile in SAS, about 7 in > StormC on my A4000/040 (33mhz), 112 MB). > > Another reason I use Storm is for WarpOS support... er, even though I > don't have a PPC card, but people I know do :) > > VBCC and GCC I've used in the past without fault, although I just can't > stand makefiles, such a pain in the ass to keep the dependances > up-to-date when you have about 50 header files... > > -- > Bart King > Developer > http://www.bartman.demon.co.uk (not up yet - soon) > > > ------------------------------------------------------------------------ > PERFORM CPR ON YOUR APR! > Get a NextCard Visa, in 30 seconds! Get rates as low as > 0.0% Intro or 9.9% Fixed APR and no hidden fees. > Apply NOW! > http://click.egroups.com/1/2121/0/_/451227/_/953066884/ > ------------------------------------------------------------------------ -- \|/ (o O) ---oOO(_)OOo--- Peekaboo!!! _____ _____ @ENDNODE @NODE "Mail_148" " FindToolType - Illegal argument... " From: Alexander Niven-Jenkins Date: 03-15-00 08:37:16 Subject: @{b}FindToolType - Illegal argument...@{ub} From: Alexander Niven-Jenkins Hiya, OK I've finally decided to scrap HiSOFT C++ and want to move to StormC (as I got it off the developer CD). Now I'm trying to compile my programs in StormC and all seems fine with no alterations needed so far except that my tooltype code has died as I get illegal argument errors, just like I used to with HiSOFT C++ when I moved to the new developer CD. char* task = NULL; BPTR olddir = NULL; struct DiskObject* dobj = NULL; // get configuration from tooltypes olddir = CurrentDir( wb->sm_ArgList[ 0 ].wa_Lock ); if( dobj = GetDiskObject( wb->sm_ArgList[ 0 ].wa_Name ) ) { if( dobj->do_ToolTypes ) { if( task = FindToolType( dobj->do_ToolTypes, TT_DELAY ) ) { Basically it chucks up an error on the if( task = FindToolType( dobj... line placing the cursor on the ',' after dobj->do_ToolTypes, so what's wrong with this? I have tried casting it to ( CONST_STRPTR ) which was the stop gap solution with HiSOFT C++ before Olaf informed me to alter exec/types.h (BTW I have reinstalled the unmodified version again). Help please? Also which libraries etc does StormC open etc within it's standard startup code? Kind regards... Alex -- Cremlin Software - http://www.cremlinsoftware.org Founder, lead programmer - ghandi@cremlinsoftware.org Production, programmer - anj@cremlinsoftware.org ------------------------------------------------------------------------ DON'T HATE YOUR RATE! Get a NextCard Visa, in 30 seconds! Get rates as low as 0.0% Intro or 9.9% Fixed APR and no hidden fees. Apply NOW! http://click.egroups.com/1/2120/0/_/451227/_/953105887/ ------------------------------------------------------------------------ @ENDNODE @NODE "Mail_149" " Re: What compiler.... " From: Steven Dobbs Date: 03-15-00 08:38:24 Subject: @{b}Re: What compiler....@{ub} From: Steven Dobbs Hello Laura On 15-Mar-00, you wrote: > From: Laura Vance > Ok, this is getting old. Every time that this same argument comes up, no > matter if it's in a mailing list, in a usenet newsgroup, or anywhere else, > it's always the same. Nobody is going to change anyone's mind about any > compiler, and odds are that whoever asked the original question just > wanted the list of compilers to make their own decision. > Laura Well, I use StormC3.0 and I have never had it working satisfactorily. For the PPC for example, asin(0.8) will give a rediculous answer such as 24. It creates code that crashes for no obvious reason. I for one would be very interested to know if there is a PPC compiler that works! Any compiler can do Hello_World.c but give them anything interesting to do and a project turns into a nightmare. vbcc seems to work quite well, it does at least have working asin/acos functions. Regards -- Steven Dobbs "AMIGA Farm" http://web.ukonline.co.uk/donkeedobbs ------------------------------------------------------------------------ GET A NEXTCARD VISA, in 30 seconds! Get rates as low as 0.0% Intro APR and no hidden fees. Apply NOW! http://click.egroups.com/1/975/0/_/451227/_/953105987/ ------------------------------------------------------------------------ @ENDNODE @NODE "Mail_150" " GRELxxxx gadget borders " From: andrewmarkwell@ukonline.co.uk Date: 03-15-00 10:55:58 Subject: @{b}GRELxxxx gadget borders@{ub} From: andrewmarkwell@ukonline.co.uk hi, I have a strgclass BOOPSI gadget (with a GA_RelWidth tag set), I have a frameiclass border too... the problem is when the window is resized... the gadget resizes ok, but the border just redraws it's old position. I have an IBox as an argument to GA_Bounds and GA_RelSpecial is set too. The border doesn't appear to size itself with respect to the string gadget, like it does for other gadgets. Andrew Markwell ------------------------------------------------------------------------ DON'T HATE YOUR RATE! Get a NextCard Visa, in 30 seconds! Get rates as low as 0.0% Intro or 9.9% Fixed APR and no hidden fees. Apply NOW! http://click.egroups.com/1/2120/0/_/451227/_/953114160/ ------------------------------------------------------------------------ @ENDNODE @NODE "Mail_151" " RE: RE: What Compiler " From: roderick_maclean@agilent.com Date: 03-15-00 11:24:26 Subject: @{b}RE: RE: What Compiler@{ub} From: roderick_maclean@agilent.com >»Funny you mention this "MakeFile". >» >»At some time 13+ years ago, when I started using Aztec Manx C, >»I encountered the "Make" program, unfortunately the syntax it used >»made no sence to me at the time, it seemed a bizare way of >»process organising on an Amiga, So, I just made up a batch file >»to do the compile, assemble & link. [snip] > I havn't seen Storm C properly but I must say I would prefer > some sort of GUI > makefile system over command line based make. Have either of you tried any of the programs which automatically create and update makefiles? IIRC they are mkmk and mkmake, you'll probably get them on Aminet - if not then a modified version of the unix mkmk command will probably do the trick (change the compiler/linker names) I think I got a version of mkmk with SAS/C but I can't remember because it was years ago. Rod -- Rod MacLean | Roderick_MacLean@agilent.com Developer - Agilent Technologies | +44 (0131) 331-7304 ** My views do not necessarily reflect the views of Agilent! ** ------------------------------------------------------------------------ PERFORM CPR ON YOUR APR! Get a NextCard Visa, in 30 seconds! Get rates as low as 0.0% Intro or 9.9% Fixed APR and no hidden fees. Apply NOW! http://click.egroups.com/1/2121/0/_/451227/_/953115886/ ------------------------------------------------------------------------ @ENDNODE @NODE "Mail_152" " Re: What compiler.... " From: Hynek Schlawack Date: 03-15-00 11:56:31 Subject: @{b}Re: What compiler....@{ub} From: Hynek Schlawack Am 15-Mär-00 schrieb Volker Barthelmann: >> Sorry but in the docs I always read, that the current version isn't >> finished pp...I'm giving vbcc a chance right now, but I don't believe >> that he will bring me away from DICE. > Well..I think the documentation of vbcc 0.7 only says that the optimizer > is not finished (something DICE doesn't have at all). There's even an item > about the version number in the FAQ. :-) Yes I know :-). > Indeed, I do not consider vbcc finished yet, because it does not yet have > all the features I'd like it to have (the other compilers don't have > these features either). However, this does not mean it's less complete > or usable - many people seem to use it quite succesfully. That's true, I've recompiled my current project with vbcc and I had not too many problems (I even learnedm, that a "case" don't open a new scope...*Sick*...but I have two problems: vbcc seems not to replace existing Exe-Files!? Before I found that out, I have compiled the whole project aprox. 5 times completly... but now he replaces it...hmm...that really confuses me (Yes, I have set the objectname using -o and no, there is no a.out). And the second point is, that vbcc is quiet slow, even when optimizing is off...are there any tricks to speed him up? Now I remebered a third problem..."make" is crashing after linking (better: the console handler)!? Has anyone this problem too? DMake doesn't have this problem... > But I have, for example, pretty precise targets regarding optimizations. > vbcc 0.7 only implements part of the optimizations which had been planned > - e.g. on a CyberstormPPC vbcc 0.7 gets about 400000 dhrystones/s whereas > my current version gets about 730000 (measured by Frank Wille). So, the > "finishedness" relates more to my own private goals rather than > completeness or stability. I found that out and I'm using it currently...until I receive this damn SAS/C from anywhere :-). Gruss, -- Hynek Schlawack ++ 7Wire Technologies ++ http://me.in-berlin.de/~hys ------------------------------------------------------------------------ PERFORM CPR ON YOUR APR! Get a NextCard Visa, in 30 seconds! Get rates as low as 0.0% Intro or 9.9% Fixed APR and no hidden fees. Apply NOW! http://click.egroups.com/1/2121/0/_/451227/_/953119075/ ------------------------------------------------------------------------ @ENDNODE @NODE "Mail_153" " Re: Re: What compiler.... " From: Michal Bartczak Date: 03-15-00 12:51:23 Subject: @{b}Re: Re: What compiler....@{ub} From: Michal Bartczak > I found that out and I'm using it currently...until I receive this damn > SAS/C from anywhere :-). I have original SasC (bought few years ago...), with instructions and so on.. But it isn't really worth it, and code it produces (ppc version) is a crap. Use vbcc ^^ Michal ------------------------------------------------------------------------ MAXIMIZE YOUR CARD, MINIMIZE YOUR RATE! Get a NextCard Visa, in 30 seconds! Get rates as low as 0.0% Intro or 9.9% Fixed APR and no hidden fees. Apply NOW! http://click.egroups.com/1/2122/0/_/451227/_/953120828/ ------------------------------------------------------------------------ @ENDNODE @NODE "Mail_154" " Re: What compiler.... " From: Hynek Schlawack Date: 03-15-00 13:24:45 Subject: @{b}Re: What compiler....@{ub} From: Hynek Schlawack Am 15-Mär-00 schrieb Michal Bartczak: >> I found that out and I'm using it currently...until I receive this damn >> SAS/C from anywhere :-). > I have original SasC (bought few years ago...), with instructions and so > on.. Oh, so you don't need it anymore :-)? > Use vbcc ^^ One big problem: Wyh isn't possible to break vc? :-((( It really sucks to have wait until a whole make is finished when you relalize, that you have forgit something :-/. Gruss, -- Hynek Schlawack ++ 7Wire Technologies ++ http://me.in-berlin.de/~hys ------------------------------------------------------------------------ DON'T HATE YOUR RATE! Get a NextCard Visa, in 30 seconds! Get rates as low as 0.0% Intro or 9.9% Fixed APR and no hidden fees. Apply NOW! http://click.egroups.com/1/2120/0/_/451227/_/953123075/ ------------------------------------------------------------------------ @ENDNODE @NODE "Mail_155" " Re: Re: What compiler.... " From: Michal Bartczak Date: 03-15-00 13:37:54 Subject: @{b}Re: Re: What compiler....@{ub} From: Michal Bartczak > >> I found that out and I'm using it currently...until I receive this damn > >> SAS/C from anywhere :-). > > I have original SasC (bought few years ago...), with instructions and so > > on.. > > Oh, so you don't need it anymore :-)? Hmm.... I think i don't ^^ > > Use vbcc ^^ > > One big problem: Wyh isn't possible to break vc? :-((( It really sucks to > have wait until a whole make is finished when you relalize, that you have > forgit something :-/. it is possible, but only when vc does output. [reboot is usually faster, when optimising ;P] It could be enough (and nice too) just to show ie. '1% done' on console ;) from time to time... Michal ------------------------------------------------------------------------ GET A NEXTCARD VISA, in 30 seconds! Get rates as low as 0.0% Intro or 9.9% Fixed APR and no hidden fees. Apply NOW! http://click.egroups.com/1/937/0/_/451227/_/953123654/ ------------------------------------------------------------------------ @ENDNODE @NODE "Mail_156" " Re: What compiler.... " From: Hynek Schlawack Date: 03-15-00 14:58:49 Subject: @{b}Re: What compiler....@{ub} From: Hynek Schlawack Am 15-Mär-00 schrieb Michal Bartczak: >> >> I found that out and I'm using it currently...until I receive this >> >> damn SAS/C from anywhere :-). >> > I have original SasC (bought few years ago...), with instructions and >> > so on.. >> Oh, so you don't need it anymore :-)? > Hmm.... I think i don't ^^ >>> Use vbcc ^^ >> One big problem: Wyh isn't possible to break vc? :-((( It really sucks to >> have wait until a whole make is finished when you relalize, that you have >> forgit something :-/. > it is possible, but only when vc does output. [reboot is usually faster, > when optimising ;P] That's exactly what I mean :-(. > It could be enough (and nice too) just to show ie. '1% done' on console ;) > from time to time... That's right...Volker? ;-) Gruss, -- Hynek Schlawack ++ 7Wire Technologies ++ http://me.in-berlin.de/~hys ------------------------------------------------------------------------ DON'T HATE YOUR RATE! Get a NextCard Visa, in 30 seconds! Get rates as low as 0.0% Intro or 9.9% Fixed APR and no hidden fees. Apply NOW! http://click.egroups.com/1/2120/0/_/451227/_/953128846/ ------------------------------------------------------------------------ @ENDNODE @NODE "Mail_157" " Re: FindToolType - Illegal argument... " From: Christian Hattemer Date: 03-15-00 16:25:07 Subject: @{b}Re: FindToolType - Illegal argument...@{ub} From: Christian Hattemer On 15-Mär-00 Alexander Niven-Jenkins wrote: > Basically it chucks up an error on the > if( task = FindToolType( dobj... StormC does a more pedantic typecheck then other compilers, which is a good thing IMHO, but you may need some additional casts to compile soures with StormC. This line is working for me: ProjfilePtr = FindToolType((CONST_STRPTR *)DObj->do_ToolTypes, "PROJECT"); > line placing the cursor on the ',' after dobj->do_ToolTypes, so what's > wrong with this? I have tried casting it to ( CONST_STRPTR ) which was the > stop gap solution with HiSOFT C++ before Olaf informed me to alter > exec/types.h (BTW I have reinstalled the unmodified version again). In which way did you alter types.h? The reason the cast for FindToolType() was needed with the old includes is that do_ToolTypes was char ** and FindToolType() expected a UBYTE **. With the new includes do_ToolTypes is STRPTR * and FindToolType() wants CONST_STRPTR *. STRPTR is unsigned char * in ANSI C mode, so is that change removing the unsigned keyword? You still can't use FindToolType() without cast. I'm also not really happy with the new layout of the OS 3.5 includes. The old files in proto/ used to include one file from clib/ and one from pragmas/. You could easily step along the sources with GoldED, form your own file via the protos file to the clib file to look at a function prototype. Now only the pragmas file is left and the include for the clib file is hidden in all the #define stuff in the pragmas file. I also added #include in proto/exec.h. It was there in the old includes and is needed to compile without warnings. Bye, Chris -- __ /// __ /// Amiga 4000T 040/40 Team *AMIGA* \\\/// \XX/ http://www.riednet.wh.tu-darmstadt.de/~chris/ It's not a bug. It's a feature. ------------------------------------------------------------------------ DON'T HATE YOUR RATE! Get a NextCard Visa, in 30 seconds! Get rates as low as 0.0% Intro or 9.9% Fixed APR and no hidden fees. Apply NOW! http://click.egroups.com/1/2120/0/_/451227/_/953135529/ ------------------------------------------------------------------------ @ENDNODE @NODE "Mail_158" " Re: What compiler.... " From: Allan Odgaard Date: 03-15-00 17:20:52 Subject: @{b}Re: What compiler....@{ub} From: Allan Odgaard On 15-Mar-00, Laura Vance wrote: > Ok, this is getting old. Every time that this same argument comes up, no > matter if it's in a mailing list, in a usenet newsgroup, or anywhere else, > it's always the same. Nobody is going to change anyone's mind about any > compiler [...] I don't see this as a stupid flame war with people saying "x is the best", but a nice insight in the various compilers strengths and weaknesses, for those of us who hasn't had the time or money to tst all of them -- so far I'd say that every recommendation has been followed by good arguments, so it's not useless. Regards Allan ------------------------------------------------------------------------ DON'T HATE YOUR RATE! Get a NextCard Visa, in 30 seconds! Get rates as low as 0.0% Intro or 9.9% Fixed APR and no hidden fees. Apply NOW! http://click.egroups.com/1/2120/0/_/451227/_/953140948/ ------------------------------------------------------------------------ @ENDNODE @NODE "Mail_159" " Re: What compiler.... " From: Allan Odgaard Date: 03-15-00 17:25:11 Subject: @{b}Re: What compiler....@{ub} From: Allan Odgaard On 15-Mar-00, Volker Barthelmann wrote: >> Anyway that's not what I disagree with most, StormC produce the fastest code >> I've seen from any Amiga compiler [...] > I think it's safe to say that if your code is fastest with StormC that's > definitely not the rule. I can second that -- my test showed the latest public version of VBCC to be 15% faster than a one year old beta of StormC V4. Since then Michael Rock told me that he'd spend a lot of time on the optimizer, but I haven't received any betas for the last year :-( Regards Allan ------------------------------------------------------------------------ GET A NEXTCARD VISA, in 30 seconds! Get rates as low as 2.9% Intro or 9.9% Fixed APR and no hidden fees. Apply NOW! http://click.egroups.com/1/936/0/_/451227/_/953140949/ ------------------------------------------------------------------------ @ENDNODE @NODE "Mail_160" " Re: What compiler.... " From: Allan Odgaard Date: 03-15-00 18:20:46 Subject: @{b}Re: What compiler....@{ub} From: Allan Odgaard On 14-Mar-00, Hynek Schlawack wrote: >> Let me start by apologizing for spending so much bandwidth on defending >> StormC. I'm just tied of seeing [...] > I developed using StormC 3.0 for aprox. one year. I think *that* is enough. Then I'm sorry I addressed you for this ranting. Your letter was just the one that filled my coup, as I often see people (users) complain about products to everybody else than the author of the particular program. > A friend of mine has reported more than enough bugs and the only answer he > got was that they will it fix asap...but they didn't. Bert Schulzki is his > name if you remember him. [...] No sorry, haven't seen his name. I too have reported dozens or more bugs, and this has resulted in many bugfixes and an invitation to their beta program, so maybe I'm not completely unbiased... also, I'd like to add that even though they have fixed a lot of bugs there still is some minor bugs pending :-( > produced and has pointed out directly the errors StormC made. But when was > the last update? H&P doing AmigaOS has unfortunately stolen nearly all the time from StormC, which is really a shame. But I gues a mainstream product as an OS upgrade pays better than a developer environment for a commercially dead platform :-( > Where is STL??? TBA I gues... to change the subject a little: I haven't really worked with STL myself, but to me it seems strange that it's a problem to port it to Amiga, cause isn't it just a lot of classes and some algorithm implementations? I.e. a bunch of headers and a link library? Or did I miss something? >> Also, I don't even think SAS/C supports 64 bit arithmetic, so if your > 64 bit isn't the only problem. The shell of StormC is crashing all the time > a.s.o. Was that with StormC 2 on a MK-III? Personally I have only experienced the shell crash once, and that was while I was borrowing an MK-III and pressing ctrl up/down without an editor window open (this normally jumps to next/prev compiler error), though it wasn't reproducible and it hasn't happend in the later versions, though my friend (who still use V2 and an MK-III) says it still happens for him. Regards Allan ------------------------------------------------------------------------ PERFORM CPR ON YOUR APR! Get a NextCard Visa, in 30 seconds! Get rates as low as 0.0% Intro or 9.9% Fixed APR and no hidden fees. Apply NOW! http://click.egroups.com/1/2121/0/_/451227/_/953140954/ ------------------------------------------------------------------------ @ENDNODE @NODE "Mail_161" " Re: FindToolType - Illegal argument... " From: Alexander Niven-Jenkins Date: 03-15-00 18:26:55 Subject: @{b}Re: FindToolType - Illegal argument...@{ub} From: Alexander Niven-Jenkins Hiya Christian, On 15-Mar-00, you wrote: >> Basically it chucks up an error on the >> if( task = FindToolType( dobj... > StormC does a more pedantic typecheck then other compilers, which is a > good thing IMHO, but you may need some additional casts to compile soures > with StormC. I know I was using HiSOFT C++ which is probably the MOST strict compiler I've seen, however it worked fine without casts. Anyway thanks that works. I was trying to cast to ( CONST_STRPTR ) which is what I did in HiSOFT C++. >> line placing the cursor on the ',' after dobj->do_ToolTypes, so what's >> wrong with this? I have tried casting it to ( CONST_STRPTR ) which was >> the stop gap solution with HiSOFT C++ before Olaf informed me to alter >> exec/types.h (BTW I have reinstalled the unmodified version again). > In which way did you alter types.h? Because HiSOFT C++ is soooo strict (although both OB and myself were unsure if it should react like this), the new includes are defined with the CONST_STRPTR, HiSOFT C++ wanted all the parameters to be const as well, which mant having to cast huge amounts of variables to be const. Olaf said just to comment out the const part of the the CONST_STRPTR typedef so it was as it was pre OS3.5. But not using HiSOFT C++ anymore I've set it back, although it might be wise to comment it out again, as I think I'll be suffering from the same problems again otherwise. *>STRPTR is unsigned char * in ANSI C mode, so is that > change removing the unsigned keyword? You still can't use FindToolType() > without cast. I could in HiSOFt C++? Kind regards... Alex -- Cremlin Software - http://www.cremlinsoftware.org Founder, lead programmer - ghandi@cremlinsoftware.org Production, programmer - anj@cremlinsoftware.org ------------------------------------------------------------------------ GET A NEXTCARD VISA, in 30 seconds! Get rates as low as 0.0% Intro or 9.9% Fixed APR and no hidden fees. Apply NOW! http://click.egroups.com/1/937/0/_/451227/_/953141227/ ------------------------------------------------------------------------ @ENDNODE @NODE "Mail_162" " Re: What compiler.... " From: Gabriele Svelto Date: 03-15-00 18:33:48 Subject: @{b}Re: What compiler....@{ub} From: Gabriele Svelto Hi Colin > Hi all, > I have been using Aztec-C (K&R) for some 13 years now. > > It still works fine (Aztec v5.2) but it's getting a bit dated, > as it doesn't have a lot of features of more modern compilers. > > I would like to find out what everyone is using & what is > prefered these days. > > Also, where do you get them from... ??? OK, seems yo look for different opinions so here's mine: I use mainly vbcc (for C code obviously) because it produces great code and it is damn easy to install & use (I tried many and many times installing GCC whithout success... :-( ). It can even produce code for PowerPC. It hasn't got an IDE though you can use it easily with GoldEd (up to the 4.7.3 version) or GoldEd Studio 6 (version 6+), for a debugging suite get MMULib.lha from Aminet. On the other hand it doesn't support C++. StormC (version 3.0 in the Developer CD) has got a quite good IDE and I love the source level debugger (great!) but it cannot compile code for ppc (there is an upgrade for that but it costs over 400DM!). I use even Dice 'cos it compiles code really faaast (I can notice the difference whithout a timer ;-) ). If you really want some advice: try them all or even use them all if you can! Ciao! Gabriele ------------------------------------------------------------------------ MAXIMIZE YOUR CARD, MINIMIZE YOUR RATE! Get a NextCard Visa, in 30 seconds! Get rates as low as 0.0% Intro or 9.9% Fixed APR and no hidden fees. Apply NOW! http://click.egroups.com/1/2122/0/_/451227/_/953142625/ ------------------------------------------------------------------------ @ENDNODE @NODE "Mail_163" " StormC and 68000... " From: Alexander Niven-Jenkins Date: 03-15-00 18:43:29 Subject: @{b}StormC and 68000...@{ub} From: Alexander Niven-Jenkins Hiya, I know that StormC produces large code due to the C lib etc, and I know that stormamiga.lib will reduce things etc, but that is 68020+ so what about a 68000 do I have to put up with the larger executable? My program has just jumped from 6k to 40k :-) Kind regards... Alex -- Cremlin Software - http://www.cremlinsoftware.org Founder, lead programmer - ghandi@cremlinsoftware.org Production, programmer - anj@cremlinsoftware.org ------------------------------------------------------------------------ GET A NEXTCARD VISA, in 30 seconds! Get rates as low as 2.9% Intro or 9.9% Fixed APR and no hidden fees. Apply NOW! http://click.egroups.com/1/936/0/_/451227/_/953142254/ ------------------------------------------------------------------------ @ENDNODE @NODE "Mail_164" " Re: STL (Was: What compiler...) " From: Alexander Niven-Jenkins Date: 03-15-00 19:34:11 Subject: @{b}Re: STL (Was: What compiler...)@{ub} From: Alexander Niven-Jenkins Hiya Allan, On 15-Mar-00, you wrote: >> Where is STL??? > TBA I gues... to change the subject a little: I haven't really worked with > STL myself, but to me it seems strange that it's a problem to port it to > Amiga, cause isn't it just a lot of classes and some algorithm > implementations? I.e. a bunch of headers and a link library? Or did I miss > something? Pretty much, but if you go to somewhere like http://www.stlport.org/ and see how much work is involved in porting the SGI STL to work on the majority of the PC C++ compilers, it's huge, I've never seen some many header files full of defines in my life (mbs of the buggers). Also I don't think the current crop of Amiga compilers support all the needed options fully. Although I guess something like vbcc/gcc will. Maybe StormC 4 will be better, it should be out in a week or 2 according to H+P (1st quarter 2000) :-) Kind regards... Alex -- Cremlin Software - http://www.cremlinsoftware.org Founder, lead programmer - ghandi@cremlinsoftware.org Production, programmer - anj@cremlinsoftware.org ------------------------------------------------------------------------ GET A NEXTCARD VISA, in 30 seconds! Get rates as low as 2.9% Intro or 9.9% Fixed APR and no hidden fees. Apply NOW! http://click.egroups.com/1/936/0/_/451227/_/953145260/ ------------------------------------------------------------------------ @ENDNODE @NODE "Mail_165" " Re: What compiler.... " From: Hynek Schlawack Date: 03-15-00 20:03:03 Subject: @{b}Re: What compiler....@{ub} From: Hynek Schlawack Am 15-Mär-00 schrieb Allan Odgaard: >>> Let me start by apologizing for spending so much bandwidth on defending >>> StormC. I'm just tied of seeing [...] >> I developed using StormC 3.0 for aprox. one year. I think *that* is >> enough. > Then I'm sorry I addressed you for this ranting. Your letter was just the > one that filled my coup, as I often see people (users) complain about > products to everybody else than the author of the particular program. That might be true. But all bugs I'm complaing about has already been reported in the ML. >> A friend of mine has reported more than enough bugs and the only answer >> he got was that they will it fix asap...but they didn't. Bert Schulzki is >> his name if you remember him. [...] > No sorry, haven't seen his name. I too have reported dozens or more bugs, > and this has resulted in many bugfixes and an invitation to their beta > program, so maybe I'm not completely unbiased... also, I'd like to add > that even though they have fixed a lot of bugs there still is some minor > bugs pending :-( There are quiet much bugs pending...also whta I hated was this buggy program that was intended to open a Screen for StormC...he didn't, if GoldEd has run....and I could not debug it as I don't have the last StormWizard (that sucks most! I wrote one program and I will never again!) headers to compile it. And an another thing is that it is not supported to use an other editor than this crapy StormEd or GoldEd. Although I've used GoldEd quiet a while, I don't want to miss CygnusEd anymore. >> produced and has pointed out directly the errors StormC made. But when >> was the last update? > H&P doing AmigaOS has unfortunately stolen nearly all the time from > StormC, which is really a shame. But I gues a mainstream product as an OS > upgrade pays better than a developer environment for a commercially dead > platform :-( Sorry, but I would say, that the main work has been done by Olaf and H&P has only coordinated the whole project. But I may be wrong however. >> Where is STL??? > TBA I gues... to change the subject a little: I haven't really worked with > STL myself, but to me it seems strange that it's a problem to port it to > Amiga, cause isn't it just a lot of classes and some algorithm > implementations? I.e. a bunch of headers and a link library? Or did I miss > something? IMHO you missed nothing...but they seem not to have the time because some STL functions are really complex. >>> Also, I don't even think SAS/C supports 64 bit arithmetic, so if your >> 64 bit isn't the only problem. The shell of StormC is crashing all the >> time a.s.o. > Was that with StormC 2 on a MK-III? Personally I have only experienced the No! It was StormC 3 on a B1240/40. > shell crash once, and that was while I was borrowing an MK-III and > pressing ctrl up/down without an editor window open (this normally jumps > to next/prev compiler error), though it wasn't reproducible and it hasn't > happend in the later versions, though my friend (who still use V2 and an > MK-III) says it still happens for him. The running shell is very unstable the rest is okay iirc. I worked a long time not with StormC anymore because it was quiet embarassing. DICE, or vbcc are much more straight-forward, but what I liked on StormC was that the compiler was quiet fast (except of modules containing complex MUI trees...). Gruss, -- Hynek Schlawack ++ 7Wire Technologies ++ http://me.in-berlin.de/~hys ------------------------------------------------------------------------ GET A NEXTCARD VISA, in 30 seconds! Get rates as low as 0.0% Intro APR and no hidden fees. Apply NOW! http://click.egroups.com/1/975/0/_/451227/_/953147801/ ------------------------------------------------------------------------ @ENDNODE @NODE "Mail_166" " Re: StormC and 68000... " From: Hynek Schlawack Date: 03-15-00 20:06:55 Subject: @{b}Re: StormC and 68000...@{ub} From: Hynek Schlawack Hello Alexander On 15-Mär-00, you wrote: > I know that StormC produces large code due to the C lib etc, and I know > that stormamiga.lib will reduce things etc, but that is 68020+ so what > about a 68000 do I have to put up with the larger executable? > My program has just jumped from 6k to 40k :-) Do think it makes sence today to support the MC68000? I think everybody who really uses his Amiga has at least an MC68030, so supporting MC68020 should be enough. Regards, -- Hynek Schlawack ++ 7Wire Technologies ++ http://me.in-berlin.de/~hys ------------------------------------------------------------------------ MAXIMIZE YOUR CARD, MINIMIZE YOUR RATE! Get a NextCard Visa, in 30 seconds! Get rates as low as 0.0% Intro or 9.9% Fixed APR and no hidden fees. Apply NOW! http://click.egroups.com/1/2122/0/_/451227/_/953147801/ ------------------------------------------------------------------------ @ENDNODE @NODE "Mail_167" " Re: Re: What compiler.... " From: Shane O'Neill Date: 03-15-00 20:16:21 Subject: @{b}Re: Re: What compiler....@{ub} From: Shane O'Neill Allan Odgaard wrote: >I can second that -- my test showed the latest public version of VBCC to be 15% >faster than a one year old beta of StormC V4. Since then Michael Rock told me >that he'd spend a lot of time on the optimizer, but I haven't received any >betas for the last year :-( I haven't really used VBCC, I had a quick look at it the other day, I thought it was quite nice. Though, I had to change my source a little as I use a lot of C++ type comments. It didn't seem to like NULL as declared in one on the headers either. I think I'll stick with StormC and GCC, so I won't have to change my comments. ;) Regards, Shane. ------------------------------------------------------------------------ GET A NEXTCARD VISA, in 30 seconds! Get rates as low as 0.0% Intro or 9.9% Fixed APR and no hidden fees. Apply NOW! http://click.egroups.com/1/937/0/_/451227/_/953147868/ ------------------------------------------------------------------------ @ENDNODE @NODE "Mail_168" " Re: FindToolType - Illegal argument... " From: Allan Odgaard Date: 03-15-00 20:49:20 Subject: @{b}Re: FindToolType - Illegal argument...@{ub} From: Allan Odgaard On 15-Mar-00, Alexander Niven-Jenkins wrote: > Also which libraries etc does StormC open etc within it's standard ^^^ ^^^ --- ??? > startup code? The startup code does AFAIK only setup SysBase (exec.library). But for each reference to a standard library base the linker lib will add init/exit code which open that library (incl. MUIMaster.library with StormAmiga.lib) -- if you want to see which libraries your program actually use then enable "write map file" from the linker settings, and look at the produced .map file. However some libraries are allowed to fail, e.g. datatypes.library, locale.library etc. -- I think this applises to all those with an init code above 3, but I'm not sure how consistent this is. If you want to have other libraries opened then you can use the LibBase class from 'libbase.h', just declare a global var e.g. like this: LibBaseC CyberGfxBase("cybergraphics.library", 41); And cybergraphics library will be opened. Supply FALSE as third parameter if you don't want the program to exit on failure. Regards Allan ------------------------------------------------------------------------ PERFORM CPR ON YOUR APR! Get a NextCard Visa, in 30 seconds! Get rates as low as 0.0% Intro or 9.9% Fixed APR and no hidden fees. Apply NOW! http://click.egroups.com/1/2121/0/_/451227/_/953150120/ ------------------------------------------------------------------------ @ENDNODE @NODE "Mail_169" " Re: FindToolType - Illegal argument... " From: Allan Odgaard Date: 03-15-00 20:53:25 Subject: @{b}Re: FindToolType - Illegal argument...@{ub} From: Allan Odgaard On 15-Mar-00, Christian Hattemer wrote: >> stop gap solution with HiSOFT C++ before Olaf informed me to alter >> exec/types.h (BTW I have reinstalled the unmodified version again). > In which way did you alter types.h? I've changed the typedef for STRPTR to be 'unsigned char *' which is quite important when strings are used together with locale library (as casting char to LONG will yield another result than unsigned char to LONG for characters above 127) -- I have not installed the latest includes from the dev. CD (if it works, don't fix it ;-) ), so I don't know if this change is already there. Regards Allan ------------------------------------------------------------------------ GET A NEXTCARD VISA, in 30 seconds! Get rates as low as 2.9% Intro or 9.9% Fixed APR and no hidden fees. Apply NOW! http://click.egroups.com/1/936/0/_/451227/_/953150122/ ------------------------------------------------------------------------ @ENDNODE @NODE "Mail_170" " Re: StormC and 68000... " From: Allan Odgaard Date: 03-15-00 20:57:27 Subject: @{b}Re: StormC and 68000...@{ub} From: Allan Odgaard On 15-Mar-00, Hynek Schlawack wrote: > Do think it makes sence today to support the MC68000? [...] I always defaults to 68020 and kick 3.0 -- so far this has resulted in maybe 3 68000 requests for TextEditor.mcc and 2 kick 2.0 requests, but nothing for my other software... so I'd drop the 68K support! ;-) Regards Allan ------------------------------------------------------------------------ GET A NEXTCARD VISA, in 30 seconds! Get rates as low as 0.0% Intro APR and no hidden fees. Apply NOW! http://click.egroups.com/1/975/0/_/451227/_/953150291/ ------------------------------------------------------------------------ @ENDNODE @NODE "Mail_171" " Re: What compiler.... " From: Allan Odgaard Date: 03-15-00 21:05:26 Subject: @{b}Re: What compiler....@{ub} From: Allan Odgaard On 15-Mar-00, Shane O'Neill wrote: > I haven't really used VBCC, I had a quick look [...] I use a lot of C++ > type comments. [...] so I won't have to change my comments. ;) I think the option to allow C++ comments is '-+'. Regards Allan ------------------------------------------------------------------------ GET A NEXTCARD VISA, in 30 seconds! Get rates as low as 0.0% Intro or 9.9% Fixed APR and no hidden fees. Apply NOW! http://click.egroups.com/1/937/0/_/451227/_/953150829/ ------------------------------------------------------------------------ @ENDNODE @NODE "Mail_172" " #Pragmas & mpega.library " From: Mike Carter Date: 03-15-00 21:10:30 Subject: @{b}#Pragmas & mpega.library@{ub} From: Mike Carter Hi, I'm using SAS-C and was trying to call some commands in the mpega.library this evening. To start with the developer includes didn't seem to match up with the rest of my include: installation. It seems that the mpega_pragmas.h was in the INCLUDE:pragma/ directory but the mpega_proto.h was looking for it in the INCLUDE:pragmas/ directory. What is the difference between pragma/ and pragamas/ and why is there two differences? Is it compiler related or something?? Next the mpega_pragamas.h file seems to be different from the others in the directory (mostly OS3.1 pragamas). mpega_pragmas.h has a line for each command similar to: #pragma amicall(MPEGABase, 0x2a, MPEGA_decode_frame(a0,a1)) Where as others I've looked at have a format like: #pragma libcall DiskBase GetUnitID 1e 001 How (if needed) can I convert the first format to the second format of pragmas, because currently SAS-C is grumbling about incorrect parameters when I build my project :-( I was hoping to get this mini project finished in one evening, but there had to be at least *one* little humpto get over!! :o) Help appreciated, or a converted/working mpega include set for SAS-C M!Ke C. -- .--------------------------------- ---- - -- --------. | AMiGA 1200EZT 060/PPC@240Mhz,50Mb RAM __/\__ | | 4 HDD,DD FD,5 CD-ROMS,CDrW,ZiP250 \ o0 / bUDDa/S^D| | SCSI+IDE,56kModem,BVGfX,19"VDU /_'-_\ | `------------------------------- --- - \/ -- ------' http://fly.to/BuddasWurld ICQ: 29173597 --- End Of Forwarded Message --- -- .--------------------------------- ---- - -- --------. | AMiGA 1200EZT 060/PPC@240Mhz,50Mb RAM __/\__ | | 4 HDD,DD FD,5 CD-ROMS,CDrW,ZiP250 \ o0 / bUDDa/S^D| | SCSI+IDE,56kModem,BVGfX,19"VDU /_'-_\ | `------------------------------- --- - \/ -- ------' http://fly.to/BuddasWurld ICQ: 29173597 ------------------------------------------------------------------------ GET A NEXTCARD VISA, in 30 seconds! Get rates as low as 2.9% Intro or 9.9% Fixed APR and no hidden fees. Apply NOW! http://click.egroups.com/1/936/0/_/451227/_/953154567/ ------------------------------------------------------------------------ @ENDNODE @NODE "Mail_173" " Re: FindToolType - Illegal argument... " From: Christian Hattemer Date: 03-15-00 21:26:19 Subject: @{b}Re: FindToolType - Illegal argument...@{ub} From: Christian Hattemer On 15-Mär-00 Alexander Niven-Jenkins wrote: > I know I was using HiSOFT C++ which is probably the MOST strict I recently heard M$ VisualC doesn't allow float x = 0.0; without warning, because 0.0 is by definition a double constant which obviously doesn't fit in a float variable without conversion. > compiler I've seen, however it worked fine without casts. Anyway thanks > that works. I was trying to cast to ( CONST_STRPTR ) which is what I did > in HiSOFT C++. CONST_STRPTR * is from the function prototype, so it's exactly what the compiler is looking for... >>> line placing the cursor on the ',' after dobj->do_ToolTypes, so what's >>> wrong with this? I have tried casting it to ( CONST_STRPTR ) which was >>> the stop gap solution with HiSOFT C++ before Olaf informed me to alter >>> exec/types.h (BTW I have reinstalled the unmodified version again). >> In which way did you alter types.h? > *>STRPTR is unsigned char * in ANSI C mode, so is that >> change removing the unsigned keyword? You still can't use FindToolType() >> without cast. > I could in HiSOFt C++? Not in StormC, I tried it some time ago. Bye, Chris -- __ /// __ /// Amiga 4000T 040/40 Team *AMIGA* \\\/// \XX/ http://www.riednet.wh.tu-darmstadt.de/~chris/ Error #0081: Volume Empty: is full. ------------------------------------------------------------------------ GET A NEXTCARD VISA, in 30 seconds! Get rates as low as 0.0% Intro or 9.9% Fixed APR and no hidden fees. Apply NOW! http://click.egroups.com/1/937/0/_/451227/_/953153449/ ------------------------------------------------------------------------ @ENDNODE @NODE "Mail_174" " Editors with Syntax highlighting " From: Dominic Clifton Date: 03-15-00 21:38:50 Subject: @{b}Editors with Syntax highlighting@{ub} From: "Dominic Clifton" I was reading a message about editors and thought to myself "I wonder what editors there are now on the Amiga that support Syntax highlighting" On the PC there are loads of really good editors (I use UltraEdit myself) that require no configuration to support syntax highlighting, you just open a .c or .cpp file and BANG! There it is, all coloured in, great! I tried GoldEd but found it's UI too clumsy (trying to be like Windows on an Amiga is NOT A GOOD THING!) and wouldn't let me configure the keyboard how I'm used to.. What other editors are there out there ? I still use TurboText, it's great, small, and very quick (except for loading large files). Cheers! Dom ------------------------------------------------------------------------ PERFORM CPR ON YOUR APR! Get a NextCard Visa, in 30 seconds! Get rates as low as 0.0% Intro or 9.9% Fixed APR and no hidden fees. Apply NOW! http://click.egroups.com/1/2121/0/_/451227/_/953157971/ ------------------------------------------------------------------------ @ENDNODE @NODE "Mail_175" " Re: FindToolType - Illegal argument... " From: Christian Hattemer Date: 03-15-00 21:40:10 Subject: @{b}Re: FindToolType - Illegal argument...@{ub} From: Christian Hattemer On 15-Mär-00 Allan Odgaard wrote: > I've changed the typedef for STRPTR to be 'unsigned char *' which is quite > important when strings are used together with locale library (as casting > char to LONG will yield another result than unsigned char to LONG for > characters above 127) -- I have not installed the latest includes from the > dev. CD (if it works, don't fix it ;-) ), so I don't know if this change > is already there. Both versions define STRPTR like this: #ifdef __cplusplus typedef char * STRPTR; /* string pointer (NULL terminated) */ #else typedef unsigned char * STRPTR; /* string pointer (NULL terminated) */ #endif Why is this diffenence made for C and C++? Bye, Chris -- __ /// __ /// Amiga 4000T 040/40 Team *AMIGA* \\\/// \XX/ http://www.riednet.wh.tu-darmstadt.de/~chris/ Tip: Zwei große Pappscheiben in die oberen Ecken Ihres Fernsehschirms kleben - jetzt sehen die Nachrichtensprecher aus wie Micky Maus! ------------------------------------------------------------------------ GET A NEXTCARD VISA, in 30 seconds! Get rates as low as 0.0% Intro APR and no hidden fees. Apply NOW! http://click.egroups.com/1/975/0/_/451227/_/953153447/ ------------------------------------------------------------------------ @ENDNODE @NODE "Mail_176" " Debugging macro trick " From: Tim Hanson Date: 03-15-00 21:41:33 Subject: @{b}Debugging macro trick@{ub} From: Tim Hanson Hi all, I have a function with this prototype: void pushFunction(void (*func)()); For debugging purposes I`d like every call to it to also pass the functions name as a char *. I thought a marco like this would do the job: #define pushFunction(a) pushFunction(a,"a") unfortunately for pushFunction(function); this expands as pushFunction(function,"a"); Any ideas or am I just shafted. -- Tim Hanson, by the River,Canterbury, Kent. Amiga A1200T,`040/25,603e/160,32MB Fast,BVisionPPC, 8.4GB IDE HD,ZIP SCSI,RENO x2CD SCSI,Pace 56K Don`t waste your time by visiting http://www.sinister67.freeserve.co.uk ------------------------------------------------------------------------ GET A NEXTCARD VISA, in 30 seconds! Get rates as low as 2.9% Intro or 9.9% Fixed APR and no hidden fees. Apply NOW! http://click.egroups.com/1/936/0/_/451227/_/953157069/ ------------------------------------------------------------------------ @ENDNODE @NODE "Mail_177" " A method to define a standard global set of text editing and selecting keyboard shortcuts is needed!! " From: Dominic Clifton Date: 03-15-00 21:52:43 Subject: @{b}A method to define a standard global set of text editing and selecting keyboard shortcuts is needed!!@{ub} From: "Dominic Clifton" Right.. I think *all* computers should have a SINGLE place to define a standard global set of text editing keyboard shortcuts. For example, in the world of Windows everyone knows that pressing shift and then cursor right will select that character and highlight it, similarly pressing control and cursor right will move the cursor to the beginning of the next word. On the Amiga (and linux too) the above operations have different results depending on what you're doing, e.g. string gadget, text editor, shell, gfx program, word processor etc.. What I want is another "Prefs" tool (and under windows on the PC it would be a control panel icon) that gives me a list of things I can assign keys to (e.g. next word, page up, page down, top of document, end of line) and a place to assign a key to it. Then I want it to save this information somewhere (ENVARC:sys/editorkeys.prefs) and then I want all new amiga programs to support it! Also, what would be nice is the ability to define what ascii characters make up white-space and what doesn't, so than when I press ctrl-left it doesn't stop/doesn't skip when I don't want it to (on a hyponated word for example). but this should be more application/profile orientated (as when editing source code you might want a different set of characters to be considered white-space) Additionally it's be nice if you could switch between different editing keyboard shortcut profiles (via a commodity?) Here's hoping for computers that get easier to use and more customizable to individual's tastes.. Dom ------------------------------------------------------------------------ MAXIMIZE YOUR CARD, MINIMIZE YOUR RATE! Get a NextCard Visa, in 30 seconds! Get rates as low as 0.0% Intro or 9.9% Fixed APR and no hidden fees. Apply NOW! http://click.egroups.com/1/2122/0/_/451227/_/953157980/ ------------------------------------------------------------------------ @ENDNODE @NODE "Mail_178" " Re: StormC and 68000... " From: Alexander Niven-Jenkins Date: 03-15-00 22:04:57 Subject: @{b}Re: StormC and 68000...@{ub} From: Alexander Niven-Jenkins Hiya Allan, On 15-Mar-00, you wrote: > I always defaults to 68020 and kick 3.0 -- so far this has resulted in > maybe 3 68000 requests for TextEditor.mcc and 2 kick 2.0 requests, but > nothing for my other software... so I'd drop the 68K support! ;-) Never really thought about it that way, even my A2000 (OS2) has an 040, I still allow 2.04 support (obviously) but 68000 seems to not be necessary then, and if need be, I could always do a special build for 68000 and those bods will have to put up with a bigger exe. Thanks! Kind regards... Alex -- Cremlin Software - http://www.cremlinsoftware.org Founder, lead programmer - ghandi@cremlinsoftware.org Production, programmer - anj@cremlinsoftware.org ------------------------------------------------------------------------ DON'T HATE YOUR RATE! Get a NextCard Visa, in 30 seconds! Get rates as low as 0.0% Intro or 9.9% Fixed APR and no hidden fees. Apply NOW! http://click.egroups.com/1/2120/0/_/451227/_/953154505/ ------------------------------------------------------------------------ @ENDNODE @NODE "Mail_179" " Re: What compiler.... (Specificly, what IDE...) " From: Dominic Clifton Date: 03-15-00 22:14:25 Subject: @{b}Re: What compiler.... (Specificly, what IDE...)@{ub} From: "Dominic Clifton" Here's my 2 cents on this... After using recent IDE's on PC system's I'm amazed that I coped for so long with the tools that I've had on my Amigas. Even borland's DOS based Turbo Pascal/Turbo C had better IDE's than *any* Amiga IDE has ever had. These days IDE's on PC's have built in GUI editors, tab-completion of function names, excellent large project management, function lookup, on-line context sensitive help (of the IDE and the C-functions/keywords/APIs etc) they're just downright brilliant to work in, everything's at your fingertips. Granted, they may not be as configurable as an Amiga (no other computer system has, or will be in my opinion) but at least the essentials for development are right there. If I had the tools on my Amiga that the PC world enjoys today I'm sure I'd still be coding on it as much as I used to, and I'd be able to spend more time coding than trying to get the damn development programs working! I've tried every amiga compiler that I've found (North-C, Manx, SAS/C, Storm, Hisoft, GCC, VBCC) and I keep trying new versions of each one, and I've never come across anything as easy to develop in than as the old DOS based Turbo C was. The nicest IDE I found on the amiga was the new Hisoft-C, but the compiler wouldn't compile most of my SAS/C specific code so I gave up on it as I didn't have the time. The main problem with Amiga IDE's (and compilers), after lack of essential features named above, is the fact that they're either not properly documented, or if they are the documentation is in another language (Mainly german). Oh well, back to my PC for a bit of Quake 3 mod making in Visual Dev Studio... MMMMmmmmMMmmmmm... Dom ------------------------------------------------------------------------ MAXIMIZE YOUR CARD, MINIMIZE YOUR RATE! Get a NextCard Visa, in 30 seconds! Get rates as low as 0.0% Intro or 9.9% Fixed APR and no hidden fees. Apply NOW! http://click.egroups.com/1/2122/0/_/451227/_/953157990/ ------------------------------------------------------------------------ @ENDNODE @NODE "Mail_180" " What PPC compiler (was: What compiler....) " From: Dominic Clifton Date: 03-15-00 22:27:20 Subject: @{b}What PPC compiler (was: What compiler....)@{ub} From: "Dominic Clifton" > I for one would be very interested to know if there is a PPC compiler that > works! Any compiler can do Hello_World.c but give them anything interesting > to do and a project turns into a nightmare. I laughed when I read this!! I have the exact same problem myself, I've tried EGCS, Storm C (PPC), and VBCC (SAS/C is out of the question because it only creates power-up and not warp-os code, PLEASE correct me if I'm wrong..) and I've never got them to compile anything fairly complicated. I managed to get a few small programs working with VBCC, but the compiler crashes randomly and so did the compiled executables, and the PPC versions of my programs turned out to be slower than the native 680x0 versions. EGCS was way to much of a pain in the ass to set up, and it wouldn't cope with most of my SAS/C code, I needed to create a great big list of #define's to stop it complaining and I didn't have the time (has anyone got any include files which make it easier to port SAS/C code to other compilers ? I'd be *eternally* grateful!) > vbcc seems to work quite well, it does at least have working asin/acos > functions. Hehe, yeah, pity it crashes on my 604/233 to damn much though. Dom p.s. Why oh why didn't I spend my £815 UK Pounds on upgrades for my PC's rather than on this damn PPC accellerator.. Argh! I should have just gone bare 060, it'd much less hassle (for all the PPC programs I use these days I might just as well). ------------------------------------------------------------------------ DON'T HATE YOUR RATE! Get a NextCard Visa, in 30 seconds! Get rates as low as 0.0% Intro or 9.9% Fixed APR and no hidden fees. Apply NOW! http://click.egroups.com/1/2120/0/_/451227/_/953157994/ ------------------------------------------------------------------------ @ENDNODE @NODE "Mail_181" " Re: What compiler.... " From: Tim Hanson Date: 03-15-00 22:39:11 Subject: @{b}Re: What compiler....@{ub} From: Tim Hanson Hi Hynek Hynek Schlawack wrote this on 14-Mar-00 at 14:27:59 > From: Hynek Schlawack > > Hello Andrzej > > On 14-Mär-00, you wrote: > >>> I never used vbcc because it seems not to be finished... >> Tell me the definition of "finished compiler" or "finished program". I >> suppose, evry soft I tell you will not be finished regarding the >> definition :-))). >> I use vbcc and I think that even in the stage of "not finished" it is >> quite great prog. > > Sorry but in the docs I always read, that the current version isn't finished > pp...I'm giving vbcc a chance right now, but I don't believe that he will > bring me away from DICE. Try compiling something that uses lots of floating point calculations and you`ll switch from dice. I still use VMake and AME from dice but any sizeable projects are compiled with vbcc. Dice makes nice small exectables though. -- Tim Hanson, by the River,Canterbury, Kent. Amiga A1200T,`040/25,603e/160,32MB Fast,BVisionPPC, 8.4GB IDE HD,ZIP SCSI,RENO x2CD SCSI,Pace 56K Don`t waste your time by visiting http://www.sinister67.freeserve.co.uk ------------------------------------------------------------------------ MAXIMIZE YOUR CARD, MINIMIZE YOUR RATE! Get a NextCard Visa, in 30 seconds! Get rates as low as 0.0% Intro or 9.9% Fixed APR and no hidden fees. Apply NOW! http://click.egroups.com/1/2122/0/_/451227/_/953163613/ ------------------------------------------------------------------------ @ENDNODE @NODE "Mail_182" " Re: What compiler.... " From: Tim Hanson Date: 03-15-00 22:52:37 Subject: @{b}Re: What compiler....@{ub} From: Tim Hanson Hi Laura Laura Vance wrote this on 15-Mar-00 at 05:10:15 > Sorry, but I get tired of the same old arguments every time this discussion > comes up. Oww! does that mean we have to stop now I was enjoying that. -------------- -- Tim Hanson, by the River,Canterbury, Kent. Amiga A1200T,`040/25,603e/160,32MB Fast,BVisionPPC, 8.4GB IDE HD,ZIP SCSI,RENO x2CD SCSI,Pace 56K Don`t waste your time by visiting http://www.sinister67.freeserve.co.uk ------------------------------------------------------------------------ GET A NEXTCARD VISA, in 30 seconds! Get rates as low as 0.0% Intro or 9.9% Fixed APR and no hidden fees. Apply NOW! http://click.egroups.com/1/937/0/_/451227/_/953163605/ ------------------------------------------------------------------------ @ENDNODE @NODE "Mail_183" " re: What compiler -- Manually talking? " From: David Erman Date: 03-15-00 23:42:19 Subject: @{b}re: What compiler -- Manually talking?@{ub} From: David Erman Hiya! Just a quick question here. If I've gotten things right here, StomrC 3.0 comes with the developer CD, right? Does that come with a printed manual as well then? I just came to think that it would diminish H&P's earnings from that product quite drastically, now wouldn't it? / Dave -- Studying Computer Sciences & Signal Processing UIN: 5361039 A4000/Apollo 040@40MHz/32Mb/1.6Gb/4X Mitsumi/M1438 RedHat Linux 6.0 -- AmigaOS 3.1 -- MacOS 8.1 ------------------------------------------------------------------------ GET A NEXTCARD VISA, in 30 seconds! Get rates as low as 2.9% Intro or 9.9% Fixed APR and no hidden fees. Apply NOW! http://click.egroups.com/1/936/0/_/451227/_/953159959/ ------------------------------------------------------------------------ @ENDNODE @NODE "Mail_184" " Re: Better editor than se in SAS/C 6.0? " From: David Erman Date: 03-15-00 23:43:09 Subject: @{b}Re: Better editor than se in SAS/C 6.0?@{ub} From: David Erman Is it just me using emacs? / Dave -- Studying Computer Sciences & Signal Processing UIN: 5361039 A4000/Apollo 040@40MHz/32Mb/1.6Gb/4X Mitsumi/M1438 RedHat Linux 6.0 -- AmigaOS 3.1 -- MacOS 8.1 ------------------------------------------------------------------------ MAXIMIZE YOUR CARD, MINIMIZE YOUR RATE! Get a NextCard Visa, in 30 seconds! Get rates as low as 0.0% Intro or 9.9% Fixed APR and no hidden fees. Apply NOW! http://click.egroups.com/1/2122/0/_/451227/_/953160009/ ------------------------------------------------------------------------ @ENDNODE @NODE "Mail_185" " Re: STL (Was: What compliers...) " From: "Nathan J. Young" Date: 03-16-00 00:07:22 Subject: @{b}Re: STL (Was: What compliers...)@{ub} From: "Nathan J. Young" Hi, > > [...] but to me it seems strange that it's a problem to port it to > > Amiga, cause isn't it just a lot of classes and some algorithm > > implementations? I.e. a bunch of headers and a link library? Or did I miss > > something? There certainly is a lot of it! The real problem is that older compilers like StormC lack a lot of the features that eventually became part of the (recent) ANSI standardisation. Especially with respect to templates, which is what the STL is composed of! > Also I don't think the current crop of Amiga compilers support all the > needed options fully. Although I guess something like vbcc/gcc will. AFAIK vbcc is only an ANSI-C compiler (not C++). I would be happy to be wrong though - I like vbcc. > Maybe StormC 4 will be better, it should be out in a week or 2 according > to H+P (1st quarter 2000) :-) I nice load of bug fixes would be appreciated. :) ttfn Nathan ------------------------------------------------------------------------ DON'T HATE YOUR RATE! Get a NextCard Visa, in 30 seconds! Get rates as low as 0.0% Intro or 9.9% Fixed APR and no hidden fees. Apply NOW! http://click.egroups.com/1/2120/0/_/451227/_/953161671/ ------------------------------------------------------------------------ @ENDNODE @NODE "Mail_186" " Re: What compiler -- Manually talking? " From: Martyn Capewell Date: 03-16-00 00:19:13 Subject: @{b}Re: What compiler -- Manually talking?@{ub} From: Martyn Capewell Hello David On 15-Mar-00, you wrote: > From: David Erman > > > Hiya! > > Just a quick question here. If I've gotten things right here, StomrC > 3.0 comes with the developer CD, right? Does that come with a printed > manual as well then? I just came to think that it would diminish H&P's > earnings from that product quite drastically, now wouldn't it? >From what I've heard, there's no manual, no PPC support, and you can't use it for non-freeware projects. I could be wrong on any or all of those points. And Storm4 is around the corner. Regards -- - Martyn Capewell ... but you can't expect to wield supreme executive power just because some watery tart threw a sword at you! ------------------------------------------------------------------------ MAXIMIZE YOUR CARD, MINIMIZE YOUR RATE! Get a NextCard Visa, in 30 seconds! Get rates as low as 0.0% Intro or 9.9% Fixed APR and no hidden fees. Apply NOW! http://click.egroups.com/1/2122/0/_/451227/_/953162393/ ------------------------------------------------------------------------ @ENDNODE @NODE "Mail_187" " Re: What compiler.... (Specificly, what IDE...) " From: Nuno Trancoso Date: 03-16-00 01:40:11 Subject: @{b}Re: What compiler.... (Specificly, what IDE...)@{ub} From: Nuno Trancoso Hya Dominic > After using recent IDE's on PC system's I'm amazed that I coped for so long > with the tools that I've had on my Amigas. Even borland's DOS based Turbo > Pascal/Turbo C had better IDE's than *any* Amiga IDE has ever had. I never used TurboC, but i do remember Turbo Pascal. It didn't had many features, AFAIKR but it's ease of use was what made it sooooo nice :)) Recently i found another niece piece of PC soft, DJGPP. Well, DJGPP is pretty much ye old gcc lookalike, so nothing new. But then there's RHIDE the IDE for DJGPP. So what do i mean with this? Well, if ppl are so desperate about an IDE why don't they program it themselves? As for myself, i don't mind using the shell :) BTW: I once took a look at MS-VC++6. I never looked again. I couldn't even manage to figure out how to write a simple "hello world" :(( The IDE was getting in my way :)) Cya l8r, DarkShadow aka Nuno Trancoso -- Infinity takes it's time. ------------------------------------------------------------------------ DON'T HATE YOUR RATE! Get a NextCard Visa, in 30 seconds! Get rates as low as 0.0% Intro or 9.9% Fixed APR and no hidden fees. Apply NOW! http://click.egroups.com/1/2120/0/_/451227/_/953255779/ ------------------------------------------------------------------------ @ENDNODE @NODE "Mail_188" " Re: What compiler -- Manually talking? " From: Nuno Trancoso Date: 03-16-00 01:45:41 Subject: @{b}Re: What compiler -- Manually talking?@{ub} From: Nuno Trancoso Hya David > Just a quick question here. If I've gotten things right here, StomrC 3.0 > comes with the developer CD, right? Does that come with a printed manual as > well then? I just came to think that it would diminish H&P's earnings from > that product quite drastically, now wouldn't it? If you don't have the manual, chances are you won't miss it much ;) (sarcastic comment) If you do get the manual, you'll have a fine time trying to decipher the sometimes cryptic "germglish", the native language of H&P. "germglish" is a strange cross between German and English but it usually doesn't "sound" right. Cya l8r, DarkShadow aka Nuno Trancoso -- Infinity takes it's time. ------------------------------------------------------------------------ PERFORM CPR ON YOUR APR! Get a NextCard Visa, in 30 seconds! Get rates as low as 0.0% Intro or 9.9% Fixed APR and no hidden fees. Apply NOW! http://click.egroups.com/1/2121/0/_/451227/_/953255779/ ------------------------------------------------------------------------ @ENDNODE @NODE "Mail_189" " Re: Re: What compiler.... " From: Shane O'Neill Date: 03-16-00 01:57:05 Subject: @{b}Re: Re: What compiler....@{ub} From: Shane O'Neill >From: Duane McDonnell >It's still annoying when people revive this topic without first >reading the archives of previous discussions which invariably will >answer their questions (as far as that's even possible with such >a subjective question). There are hundreds of archived messages >on this topic which are easily found using a search engine. It's >common courtesy to utilize these resources before starting a thread >on a mail based list which will predictably cause a large increase >in redundant traffic. Personally I have no problem with specific >questions, but open-ended ones like "which is the best compiler?" >just highlight laziness and inconsideration on the part of the >poster since it's obvious they haven't bothered to do any prior >research of their own. If you don't have the time to research >these things, don't assume other people will have the patience >to sit through yet another round of this tiresome debate. I certainly don't think there's a problem here, technology and development are dynamic and move on and change. Questions answered in archives a while back are not necessarily going to bring the same answers as that same question asked more recently. If you don't like it just ignore it, that's what I'd do. Regards, Shane. ------------------------------------------------------------------------ MAXIMIZE YOUR CARD, MINIMIZE YOUR RATE! Get a NextCard Visa, in 30 seconds! Get rates as low as 0.0% Intro or 9.9% Fixed APR and no hidden fees. Apply NOW! http://click.egroups.com/1/2122/0/_/451227/_/953168313/ ------------------------------------------------------------------------ @ENDNODE @NODE "Mail_190" " Re: What compiler.... (Specificly, what IDE...) " From: Colin Wenzel Date: 03-16-00 02:40:53 Subject: @{b}Re: What compiler.... (Specificly, what IDE...)@{ub} From: Colin Wenzel On 16-Mar-00, Dominic Clifton wrote: -snip- > These days IDE's on PC's have built in GUI editors, tab-completion of > function names, excellent large project management, function lookup, > on-line context sensitive help (of the IDE and the > C-functions/keywords/APIs etc) they're just downright brilliant to > work in, everything's at your fingertips. > Ok, I just can't stand it any longer, i'll ask................... (:/ What's "IDE" stand for ?? (Besides harddrive types) I'm partially acronymically disabled..... ------------------------------------- Colin Wenzel. Australia. EMAIL: mailto:colstv@hotkey.net.au URL: http://www.hotkey.net.au/~colstv/ ICQ: 17608330 AMIGA: 4000T, 060/50, 150Mb RAM, O.S 3.5 --------------------------------------- ------------------------------------------------------------------------ MAXIMIZE YOUR CARD, MINIMIZE YOUR RATE! Get a NextCard Visa, in 30 seconds! Get rates as low as 0.0% Intro or 9.9% Fixed APR and no hidden fees. Apply NOW! http://click.egroups.com/1/2122/0/_/451227/_/953171007/ ------------------------------------------------------------------------ @ENDNODE @NODE "Mail_191" " Re: StormC and 68000... " From: Colin Wenzel Date: 03-16-00 02:41:34 Subject: @{b}Re: StormC and 68000...@{ub} From: Colin Wenzel On 16-Mar-00, Allan Odgaard wrote: > From: Allan Odgaard > On 15-Mar-00, Hynek Schlawack wrote: >> Do think it makes sence today to support the MC68000? [...] > > I always defaults to 68020 and kick 3.0 -- so far this has resulted > in maybe 3 68000 requests for TextEditor.mcc and 2 kick 2.0 requests, > but nothing for my other software... so I'd drop the 68K support! ;-) > > Regards Allan But then technically it wouldn't be a 68xxx compiler anymore. Plus !!! What am I going to do with my A600 ??? (:> ------------------------------------- Colin Wenzel. Australia. EMAIL: mailto:colstv@hotkey.net.au URL: http://www.hotkey.net.au/~colstv/ ICQ: 17608330 AMIGA: 4000T, 060/50, 150Mb RAM, O.S 3.5 --------------------------------------- ------------------------------------------------------------------------ PERFORM CPR ON YOUR APR! Get a NextCard Visa, in 30 seconds! Get rates as low as 0.0% Intro or 9.9% Fixed APR and no hidden fees. Apply NOW! http://click.egroups.com/1/2121/0/_/451227/_/953171015/ ------------------------------------------------------------------------ @ENDNODE @NODE "Mail_192" " Re: Re: What compiler.... (Specificly, what IDE...) " From: Shane O'Neill Date: 03-16-00 02:55:44 Subject: @{b}Re: Re: What compiler.... (Specificly, what IDE...)@{ub} From: Shane O'Neill >From: Colin Wenzel >Ok, I just can't stand it any longer, i'll ask................... (:/ > >What's "IDE" stand for ?? (Besides harddrive types) >I'm partially acronymically disabled..... Integrated Development Environment. Regards, Shane. ------------------------------------------------------------------------ MAXIMIZE YOUR CARD, MINIMIZE YOUR RATE! Get a NextCard Visa, in 30 seconds! Get rates as low as 0.0% Intro or 9.9% Fixed APR and no hidden fees. Apply NOW! http://click.egroups.com/1/2122/0/_/451227/_/953171931/ ------------------------------------------------------------------------ @ENDNODE @NODE "Mail_193" " Re: Re: What compiler.... (Specificly, what IDE...) " From: Shane O'Neill Date: 03-16-00 03:21:04 Subject: @{b}Re: Re: What compiler.... (Specificly, what IDE...)@{ub} From: Shane O'Neill >From: "Dominic Clifton" >Here's my 2 cents on this... >After using recent IDE's on PC system's I'm amazed that I coped for so long with >the tools that I've had on my Amigas. Even borland's DOS based Turbo >Pascal/Turbo C had better IDE's than *any* Amiga IDE has ever had. Besides my gripes with Pascal, I really don't think that the Pascal DOS IDE is anything special. It's certinaly not better that all the Amiga IDEs I've used. >These days IDE's on PC's have built in GUI editors, tab-completion of function >names, excellent large project management, function lookup, on-line context >sensitive help (of the IDE and the C-functions/keywords/APIs etc) they're just >downright brilliant to work in, everything's at your fingertips. I've never seen a built in GUI editor I've been happy with. The best I've seen IMO are (Amiga = CEd, Windows = Textpad, Linux = Vi) Though if the Amiga had a much better market share then I'm sure we would see much better IDEs. >Granted, they may not be as configurable as an Amiga (no other computer system >has, or will be in my opinion) but at least the essentials for development are >right there. >If I had the tools on my Amiga that the PC world enjoys today I'm sure I'd still >be coding on it as much as I used to, and I'd be able to spend more time coding >than trying to get the damn development programs working! >I've tried every amiga compiler that I've found (North-C, Manx, SAS/C, Storm, >Hisoft, GCC, VBCC) and I keep trying new versions of each one, and I've never >come across anything as easy to develop in than as the old DOS based Turbo C >was. The nicest IDE I found on the amiga was the new Hisoft-C, but the compiler >wouldn't compile most of my SAS/C specific code so I gave up on it as I didn't >have the time. The main problem with Amiga IDE's (and compilers), after lack of >essential features named above, is the fact that they're either not properly >documented, or if they are the documentation is in another language (Mainly >german). >Oh well, back to my PC for a bit of Quake 3 mod making in Visual Dev Studio... >MMMMmmmmMMmmmmm... Visual Dev Studio. Yuck, Borlands C++ Builder is a better c development package IMO, still I don't like programming under windows at all. Linux and Amiga are a different matter though. Regards, Shane. ------------------------------------------------------------------------ DON'T HATE YOUR RATE! Get a NextCard Visa, in 30 seconds! Get rates as low as 0.0% Intro or 9.9% Fixed APR and no hidden fees. Apply NOW! http://click.egroups.com/1/2120/0/_/451227/_/953174287/ ------------------------------------------------------------------------ @ENDNODE @NODE "Mail_194" " Re: Better editor than se in SAS/C 6.0? " From: Kelly Samel Date: 03-16-00 03:39:56 Subject: @{b}Re: Better editor than se in SAS/C 6.0?@{ub} From: Kelly Samel Hello David On 15-Mar-00, you wrote: > From: David Erman > > > Is it just me using emacs? > > / Dave I think you are the only one left. ;) I use BED and SAS/C. I find the combination of them to work great. I have never encountered a compiler bug with SAS/C, I do not use any C++ though. I have found SAS/C to be a very robust and capable compiler when compared to many others, and I love the 'icon' based project drawer. It is easy to re-route the error messages to BED too. Also does Storm or HiSoft C have an equivalent function to stci_d()? I have been unable to find one. ------------------------------------------------------------------------ GET A NEXTCARD VISA, in 30 seconds! Get rates as low as 2.9% Intro or 9.9% Fixed APR and no hidden fees. Apply NOW! http://click.egroups.com/1/936/0/_/451227/_/953174449/ ------------------------------------------------------------------------ @ENDNODE @NODE "Mail_195" " Re: What compiler.... (Specificly, what IDE...) " From: Kelly Samel Date: 03-16-00 03:52:15 Subject: @{b}Re: What compiler.... (Specificly, what IDE...)@{ub} From: Kelly Samel Hello Dominic On 15-Mar-00, you wrote: > From: "Dominic Clifton" > After using recent IDE's on PC system's I'm amazed that I coped for so > long with the tools that I've had on my Amigas. Even borland's DOS based > Turbo Pascal/Turbo C had better IDE's than *any* Amiga IDE has ever had. > > These days IDE's on PC's have built in GUI editors, tab-completion of > function names, excellent large project management, function lookup, > on-line context sensitive help (of the IDE and the > C-functions/keywords/APIs etc) they're just downright brilliant to work > in, everything's at your fingertips. > > Granted, they may not be as configurable as an Amiga (no other computer > system has, or will be in my opinion) but at least the essentials for > development are right there. > > If I had the tools on my Amiga that the PC world enjoys today I'm sure I'd > still be coding on it as much as I used to, and I'd be able to spend more > time coding than trying to get the damn development programs working! > > I've tried every amiga compiler that I've found (North-C, Manx, SAS/C, > Storm, Hisoft, GCC, VBCC) and I keep trying new versions of each one, and > I've never come across anything as easy to develop in than as the old DOS > based Turbo C was. The nicest IDE I found on the amiga was the new > Hisoft-C, but the compiler wouldn't compile most of my SAS/C specific code > so I gave up on it as I didn't have the time. The main problem with Amiga > IDE's (and compilers), after lack of essential features named above, is > the fact that they're either not properly documented, or if they are the > documentation is in another language (Mainly german). > > Oh well, back to my PC for a bit of Quake 3 mod making in Visual Dev > Studio... MMMMmmmmMMmmmmm... > > Dom Oh, please! I used Turbo C at College for over 6 months and I think Storm and HiSoft easily compete with it. TurboC is painfully slow on a 150Mhz Pentium, Sas/C compiles at least four times faster on my 060. I did use the Windows version of TurboC, but I can not imagine the Dos version being any better. I find it much preferable to develop on the Amiga, I find it to be much easier to get into doing fun things like gfx, sound, animation ect. than PC coding. I am not saying TurboC is bad, it is fairly good, but I think the Amiga compilers are as good or better in many areas and Amiga coding is still more fun. As for features, I find programs like ArtPro and Numerator as well as many other freeware goodies fill the gaps nicely and offer more options overall than an 'all-in-one' program ever will. ------------------------------------------------------------------------ DON'T HATE YOUR RATE! Get a NextCard Visa, in 30 seconds! Get rates as low as 0.0% Intro or 9.9% Fixed APR and no hidden fees. Apply NOW! http://click.egroups.com/1/2120/0/_/451227/_/953175188/ ------------------------------------------------------------------------ @ENDNODE @NODE "Mail_196" " Re: Re: What compiler.... " From: Volker Barthelmann Date: 03-16-00 04:57:55 Subject: @{b}Re: Re: What compiler....@{ub} From: Volker Barthelmann On Wed, 15 Mar 2000, Hynek Schlawack wrote: > That's true, I've recompiled my current project with vbcc and I had not too > many problems (I even learnedm, that a "case" don't open a new > scope...*Sick*...but I have two problems: vbcc seems not to replace > existing Exe-Files!? Before I found that out, I have compiled the whole > project aprox. 5 times completly... but now he replaces it...hmm...that > really confuses me (Yes, I have set the objectname using -o and no, there > is no a.out). I don't think I've heard such a problem before. Might be caused by something else. If you're able to reproduce it, please send a bug report. > And the second point is, that vbcc is quiet slow, even when > optimizing is off...are there any tricks to speed him up? Apart from putting frontend, preprocessor, compiler, assembler and linker into RAM, no. Actually, I think vbcc compiles about as fast as most other compilers. However, if you've been used to DICE then vbcc will never be that fast. I think DICE does much less dynamic memory allocation, omits most type-checking and generally only implements a subset of ISO/ANSI C - but it's really fast. > Now I remebered a > third problem..."make" is crashing after linking (better: the console > handler)!? Has anyone this problem too? DMake doesn't have this problem... Perhaps the stack size is too small or your make has problems? Volker ------------------------------------------------------------------------ DON'T HATE YOUR RATE! Get a NextCard Visa, in 30 seconds! Get rates as low as 0.0% Intro or 9.9% Fixed APR and no hidden fees. Apply NOW! http://click.egroups.com/1/2120/0/_/451227/_/953179912/ ------------------------------------------------------------------------ @ENDNODE @NODE "Mail_197" " Re: Re: What compiler.... " From: Volker Barthelmann Date: 03-16-00 05:04:42 Subject: @{b}Re: Re: What compiler....@{ub} From: Volker Barthelmann On Wed, 15 Mar 2000, Hynek Schlawack wrote: > Am 15-Mär-00 schrieb Michal Bartczak: > >> One big problem: Wyh isn't possible to break vc? :-((( It really sucks to > >> have wait until a whole make is finished when you relalize, that you have > >> forgit something :-/. > > it is possible, but only when vc does output. [reboot is usually faster, > > when optimising ;P] > > That's exactly what I mean :-(. Ah, I see. I'm usually developing under Unix and there Ctrl-C works immediately. :-) > > It could be enough (and nice too) just to show ie. '1% done' on console ;) > > from time to time... > > That's right...Volker? ;-) I will try to improve this for the next release (though I don't like the "x% complete"-idea at all :-). If anybody has suggestions on how to implement this on AmigaOS, please send me a mail. Volker ------------------------------------------------------------------------ PERFORM CPR ON YOUR APR! Get a NextCard Visa, in 30 seconds! Get rates as low as 0.0% Intro or 9.9% Fixed APR and no hidden fees. Apply NOW! http://click.egroups.com/1/2121/0/_/451227/_/953179907/ ------------------------------------------------------------------------ @ENDNODE @NODE "Mail_198" " Re: Debugging macro trick " From: Volker Barthelmann Date: 03-16-00 05:09:21 Subject: @{b}Re: Debugging macro trick@{ub} From: Volker Barthelmann On Wed, 15 Mar 2000, Tim Hanson wrote: > For debugging purposes I`d like every call to it to also pass the functions > name as a char *. I thought a marco like this would do the job: > > #define pushFunction(a) pushFunction(a,"a") > > unfortunately for > > pushFunction(function); > > this expands as > > pushFunction(function,"a"); > > Any ideas or am I just shafted. #define str2(x) #x #define str(x) str2(x) #define pushFunction(a) pushFunction(a,str(a)) Volker ------------------------------------------------------------------------ GET A NEXTCARD VISA, in 30 seconds! Get rates as low as 0.0% Intro or 9.9% Fixed APR and no hidden fees. Apply NOW! http://click.egroups.com/1/937/0/_/451227/_/953179905/ ------------------------------------------------------------------------ @ENDNODE @NODE "Mail_199" " stci_d() (was: Re: Better editor than se in SAS/C 6.0?) " From: Allan Odgaard Date: 03-16-00 09:04:42 Subject: @{b}stci_d() (was: Re: Better editor than se in SAS/C 6.0?)@{ub} From: Allan Odgaard On 16-Mar-00, Kelly Samel wrote: > [...] Also does Storm or HiSoft C have an equivalent > function to stci_d()? I have been unable to find one. What is it? Regards Allan -- URL: http://www.diku.dk/students/duff/ OS/2 - half of operating system ------------------------------------------------------------------------ GET A NEXTCARD VISA, in 30 seconds! Get rates as low as 2.9% Intro or 9.9% Fixed APR and no hidden fees. Apply NOW! http://click.egroups.com/1/936/0/_/451227/_/953194830/ ------------------------------------------------------------------------ @ENDNODE @NODE "Mail_200" " Antwort: Re: STL (Was: What compiler...) " From: Raimund.Dold@bgt.de Date: 03-16-00 09:57:41 Subject: @{b}Antwort: Re: STL (Was: What compiler...)@{ub} From: Raimund.Dold@bgt.de STL is no problem if you use g++, I even tried to compile it for ppc and it did basically work, only streams didn't. Raimund Alexander Niven-Jenkins on 15.03.2000 19:37:47 Bitte antworten an amiga-c@onelist.com An: amiga-c@onelist.com Kopie: Thema: [amiga-c] Re: STL (Was: What compiler...) From: Alexander Niven-Jenkins Hiya Allan, On 15-Mar-00, you wrote: >> Where is STL??? > TBA I gues... to change the subject a little: I haven't really worked with > STL myself, but to me it seems strange that it's a problem to port it to > Amiga, cause isn't it just a lot of classes and some algorithm > implementations? I.e. a bunch of headers and a link library? Or did I miss > something? Pretty much, but if you go to somewhere like http://www.stlport.org/ and see how much work is involved in porting the SGI STL to work on the majority of the PC C++ compilers, it's huge, I've never seen some many header files full of defines in my life (mbs of the buggers). Also I don't think the current crop of Amiga compilers support all the needed options fully. Although I guess something like vbcc/gcc will. Maybe StormC 4 will be better, it should be out in a week or 2 according to H+P (1st quarter 2000) :-) Kind regards... Alex -- Cremlin Software - http://www.cremlinsoftware.org Founder, lead programmer - ghandi@cremlinsoftware.org Production, programmer - anj@cremlinsoftware.org ------------------------------------------------------------------------ GET A NEXTCARD VISA, in 30 seconds! Get rates as low as 2.9% Intro or 9.9% Fixed APR and no hidden fees. Apply NOW! http://click.egroups.com/1/936/0/_/451227/_/953145260/ ------------------------------------------------------------------------ ------------------------------------------------------------------------ MAXIMIZE YOUR CARD, MINIMIZE YOUR RATE! Get a NextCard Visa, in 30 seconds! Get rates as low as 0.0% Intro or 9.9% Fixed APR and no hidden fees. Apply NOW! http://click.egroups.com/1/2122/0/_/451227/_/953189723/ ------------------------------------------------------------------------ @ENDNODE @NODE "Mail_201" " How to find the Current Window? " From: David Erman Date: 03-16-00 11:26:41 Subject: @{b}How to find the Current Window?@{ub} From: David Erman Hiya! I was wondering, If I've got this background process running, which isn't attached to a screen/window in a natural way, how do I find out the current window I need for attaching my EasyRequest()'s to? / Dave -- Studying Computer Sciences & Signal Processing UIN: 5361039 A4000/Apollo 040@40MHz/32Mb/1.6Gb/4X Mitsumi/M1438 RedHat Linux 6.0 -- AmigaOS 3.1 -- MacOS 8.1 ------------------------------------------------------------------------ DON'T HATE YOUR RATE! Get a NextCard Visa, in 30 seconds! Get rates as low as 0.0% Intro or 9.9% Fixed APR and no hidden fees. Apply NOW! http://click.egroups.com/1/2120/0/_/451227/_/953202220/ ------------------------------------------------------------------------ @ENDNODE @NODE "Mail_202" " Single precision math functions " From: Gabriele Svelto Date: 03-16-00 11:55:29 Subject: @{b}Single precision math functions@{ub} From: Gabriele Svelto Hello 'verybody, does anybody know how do I do single precision math functions under VBCC in native mode (i.e. whithout using mathieee* libraries) both for 68k and vbcc processors? I know that standard ANSI functions are double precision so I cannot use them (too slow...). Thanx in advance Gabriele ------------------------------------------------------------------------ MAXIMIZE YOUR CARD, MINIMIZE YOUR RATE! Get a NextCard Visa, in 30 seconds! Get rates as low as 0.0% Intro or 9.9% Fixed APR and no hidden fees. Apply NOW! http://click.egroups.com/1/2122/1/_/451227/_/953315635/ ------------------------------------------------------------------------ @ENDNODE @NODE "Mail_203" " Re: What compiler.... " From: Hynek Schlawack Date: 03-16-00 15:04:33 Subject: @{b}Re: What compiler....@{ub} From: Hynek Schlawack Am 16-Mär-00 schrieb Volker Barthelmann: >> That's true, I've recompiled my current project with vbcc and I had not >> too many problems (I even learnedm, that a "case" don't open a new >> scope...*Sick*...but I have two problems: vbcc seems not to replace >> existing Exe-Files!? Before I found that out, I have compiled the whole >> project aprox. 5 times completly... but now he replaces it...hmm...that >> really confuses me (Yes, I have set the objectname using -o and no, there >> is no a.out). > I don't think I've heard such a problem before. Might be caused by > something else. If you're able to reproduce it, please send a bug report. I'll do but I haven'þ the problem anymore...I had changed a little bit in a MUI gui and I was wondering all the time why the gui is looking exactly the same after a Compilation :-). >> And the second point is, that vbcc is quiet slow, even when >> optimizing is off...are there any tricks to speed him up? > Apart from putting frontend, preprocessor, compiler, assembler and linker > into RAM, no. Actually, I think vbcc compiles about as fast as most other they ain't pure, richt? > compilers. However, if you've been used to DICE then vbcc will never be Not exactly, SAS/C can be with gst even faster..but yers, I have read the FAQ and I know that now precompiled headers are planed :-). > that fast. I think DICE does much less dynamic memory allocation, omits > most type-checking and generally only implements a subset of ISO/ANSI C - > but it's really fast. Yes, it's damn fast :-). >> Now I remebered a >> third problem..."make" is crashing after linking (better: the console >> handler)!? Has anyone this problem too? DMake doesn't have this >> problem... > Perhaps the stack size is too small or your make has problems? I don't think, that the stack is the problem as I have a stack of 124.000 bytes. And I'm talking about the GNU-make which hasn't crashed before even while using a stack-usage-moster like gcc :-). Gruss, -- Hynek Schlawack ++ 7Wire Technologies ++ http://me.in-berlin.de/~hys ------------------------------------------------------------------------ GET A NEXTCARD VISA, in 30 seconds! Get rates as low as 2.9% Intro or 9.9% Fixed APR and no hidden fees. Apply NOW! http://click.egroups.com/1/936/0/_/451227/_/953217828/ ------------------------------------------------------------------------ @ENDNODE @NODE "Mail_204" " Re: What compiler.... " From: Hynek Schlawack Date: 03-16-00 15:05:52 Subject: @{b}Re: What compiler....@{ub} From: Hynek Schlawack Am 16-Mär-00 schrieb Volker Barthelmann: >> >> One big problem: Wyh isn't possible to break vc? :-((( It really sucks >> >> to have wait until a whole make is finished when you relalize, that >> >> you have forgit something :-/. >> > it is possible, but only when vc does output. [reboot is usually >> > faster, when optimising ;P] >> That's exactly what I mean :-(. > Ah, I see. I'm usually developing under Unix and there Ctrl-C works > immediately. :-) I see :-). That's exactly what I'm missing on the Amiga...and in your compiler :-). >> > It could be enough (and nice too) just to show ie. '1% done' on console >> > ;) from time to time... >> That's right...Volker? ;-) > I will try to improve this for the next release (though I don't like the > "x% complete"-idea at all :-). If anybody has suggestions on how to > implement this on AmigaOS, please send me a mail. Yes, % is not a good idea, but perhaps something else...I'm sure you'll find something :-). Gruss, -- Hynek Schlawack ++ 7Wire Technologies ++ http://me.in-berlin.de/~hys ------------------------------------------------------------------------ GET A NEXTCARD VISA, in 30 seconds! Get rates as low as 2.9% Intro or 9.9% Fixed APR and no hidden fees. Apply NOW! http://click.egroups.com/1/936/0/_/451227/_/953217830/ ------------------------------------------------------------------------ @ENDNODE @NODE "Mail_205" " Re: What compiler.... " From: Hynek Schlawack Date: 03-16-00 15:12:28 Subject: @{b}Re: What compiler....@{ub} From: Hynek Schlawack Am 15-Mär-00 schrieb Tim Hanson: >>>> I never used vbcc because it seems not to be finished... >>> Tell me the definition of "finished compiler" or "finished program". I >>> suppose, evry soft I tell you will not be finished regarding the >>> definition :-))). >>> I use vbcc and I think that even in the stage of "not finished" it is >>> quite great prog. >> Sorry but in the docs I always read, that the current version isn't >> finished pp...I'm giving vbcc a chance right now, but I don't believe >> that he will bring me away from DICE. > Try compiling something that uses lots of floating point calculations and Hm, I never really used floats...strange... :-) > you`ll switch from dice. I still use VMake and AME from dice but any I never used VMake...I hate IDEs. > sizeable projects are compiled with vbcc. Dice makes nice small exectables > though. Yes, I'm currently using DICE but it's awful to wait many minutes until a damned project is compiled...especially when adding a catalog string etc. Gruss, -- Hynek Schlawack ++ 7Wire Technologies ++ http://me.in-berlin.de/~hys ------------------------------------------------------------------------ MAXIMIZE YOUR CARD, MINIMIZE YOUR RATE! Get a NextCard Visa, in 30 seconds! Get rates as low as 0.0% Intro or 9.9% Fixed APR and no hidden fees. Apply NOW! http://click.egroups.com/1/2122/0/_/451227/_/953217832/ ------------------------------------------------------------------------ @ENDNODE @NODE "Mail_206" " Re: stci_d() (was: Re: Better editor than se in SAS/C 6.0?) " From: Olaf Barthel Date: 03-16-00 15:24:20 Subject: @{b}Re: stci_d() (was: Re: Better editor than se in SAS/C 6.0?)@{ub} From: "Olaf Barthel" On Mar 16 Allan (Allan Odgaard) wrote: > From: Allan Odgaard > > On 16-Mar-00, Kelly Samel wrote: > > > [...] Also does Storm or HiSoft C have an equivalent > > function to stci_d()? I have been unable to find one. > > What is it? It converts a number into a string (this is a non-portable SAS/C function). This makes you wonder why Kelly doesn't use sprintf() instead ;) -- Home: Olaf Barthel, Brabeckstrasse 35, D-30559 Hannover Net: olsen@sourcery.han.de (Home), olsen@logicalline.com (Work) ------------------------------------------------------------------------ MAXIMIZE YOUR CARD, MINIMIZE YOUR RATE! Get a NextCard Visa, in 30 seconds! Get rates as low as 0.0% Intro or 9.9% Fixed APR and no hidden fees. Apply NOW! http://click.egroups.com/1/2122/0/_/451227/_/953243924/ ------------------------------------------------------------------------ @ENDNODE @NODE "Mail_207" " RE: #Pragmas & mpega.library " From: Jesper Svennevid Date: 03-16-00 15:42:49 Subject: @{b}RE: #Pragmas & mpega.library@{ub} From: Jesper Svennevid > From: Mike Carter > > Hi, > > I'm using SAS-C and was trying to call some commands in the > mpega.library this evening. To start with the developer includes > didn't seem to match up with the rest of my include: installation. > It seems that the mpega_pragmas.h was in the INCLUDE:pragma/ > directory but the mpega_proto.h was looking for it in the > INCLUDE:pragmas/ directory. What is the difference between pragma/ > and pragamas/ and why is there two differences? Is it compiler > related or something?? > > Next the mpega_pragamas.h file seems to be different from the others > in the directory (mostly OS3.1 pragamas). mpega_pragmas.h has a line > for each command similar to: > > #pragma amicall(MPEGABase, 0x2a, MPEGA_decode_frame(a0,a1)) > > Where as others I've looked at have a format like: > > #pragma libcall DiskBase GetUnitID 1e 001 > > How (if needed) can I convert the first format to the second format of > pragmas, because currently SAS-C is grumbling about incorrect > parameters when I build my project :-( > > I was hoping to get this mini project finished in one evening, but > there had to be at least *one* little humpto get over!! :o) > > Help appreciated, or a converted/working mpega include set for SAS-C Is there a .fd-file included in the mpega-includes? If there is, you could use fd2pragma(included in the sasc-tools) to generate a pragma- file. -- // Jesper Svennevid : jesper@fxrealm.com Software Developer at FXRealm Studios http://www.fxrealm.com/ ------------------------------------------------------------------------ GET A NEXTCARD VISA, in 30 seconds! Get rates as low as 2.9% Intro or 9.9% Fixed APR and no hidden fees. Apply NOW! http://click.egroups.com/1/936/0/_/451227/_/953231738/ ------------------------------------------------------------------------ @ENDNODE @NODE "Mail_208" " Re: Re: What compiler.... " From: Duane McDonnell Date: 03-16-00 15:51:40 Subject: @{b}Re: Re: What compiler....@{ub} From: Duane McDonnell >From: Allan Odgaard >On 15-Mar-00, Laura Vance wrote: >> Ok, this is getting old. Every time that this same argument comes up, no >> matter if it's in a mailing list, in a usenet newsgroup, or anywhere else, >> it's always the same. Nobody is going to change anyone's mind about any >> compiler [...] >I don't see this as a stupid flame war with people saying "x is the best", but >a nice insight in the various compilers strengths and weaknesses, for those of >us who hasn't had the time or money to tst all of them -- so far I'd say that >every recommendation has been followed by good arguments, so it's not useless. It's still annoying when people revive this topic without first reading the archives of previous discussions which invariably will answer their questions (as far as that's even possible with such a subjective question). There are hundreds of archived messages on this topic which are easily found using a search engine. It's common courtesy to utilize these resources before starting a thread on a mail based list which will predictably cause a large increase in redundant traffic. Personally I have no problem with specific questions, but open-ended ones like "which is the best compiler?" just highlight laziness and inconsideration on the part of the poster since it's obvious they haven't bothered to do any prior research of their own. If you don't have the time to research these things, don't assume other people will have the patience to sit through yet another round of this tiresome debate. ------------------------------------------------------------------------ MAXIMIZE YOUR CARD, MINIMIZE YOUR RATE! Get a NextCard Visa, in 30 seconds! Get rates as low as 0.0% Intro or 9.9% Fixed APR and no hidden fees. Apply NOW! http://click.egroups.com/1/2122/0/_/451227/_/953155732/ ------------------------------------------------------------------------ @ENDNODE @NODE "Mail_209" " RE: What PPC compiler (was: What compiler....) " From: Jesper Svennevid Date: 03-16-00 15:54:52 Subject: @{b}RE: What PPC compiler (was: What compiler....)@{ub} From: Jesper Svennevid > From: "Dominic Clifton" > > I laughed when I read this!! I have the exact same problem > myself, I've tried EGCS, Storm C (PPC), and VBCC (SAS/C is > out of the question because it only creates power-up and not > warp-os code, PLEASE correct me if I'm wrong..) > and I've never got them to compile anything fairly complicated. > I managed to get a few small programs working with VBCC, but > the compiler crashes randomly and so did the compiled executables, > and the PPC versions of my programs turned out to be slower > than the native 680x0 versions. I use EGCS for compiling Snes9X, and I almost never get any problems, and when I get I know how to avoid them (as for example the compiler crashes when a case is caught outside a switch :), and it's quite a lot of code, the ppc-executable becomes around 800kB in size (post-processed with elf2exe2, with elf2exe 1 it ends up around 1.3MB :) > EGCS was way to much of a pain in the ass to set up, and it > wouldn't cope with most of my SAS/C code, I needed to create > a great big list of #define's to stop it complaining and I > didn't have the time (has anyone got any include files which > make it easier to port SAS/C code to other compilers ? I'd be > *eternally* grateful!) What kind of includes? You cannot use the sas/c includes directly, as egcs doesn't handle pragmas too well :) I've built all my includes from scratch using the GG-includes (both the standard and the aos ones), processed them with geninclude.pl from Holger Jakobs homepage so that they should be properly aligned, added both the powerup and warpos-includes, generated new inline and proto-files, and added includes (slightly modified) from paladins homepage. (Do NOT use paladins includes unmodified when developing warpos, you get nasty instability problems due to a few small flaws in them.) > > vbcc seems to work quite well, it does at least have > > working asin/acos functions. > > Hehe, yeah, pity it crashes on my 604/233 to damn much though. > > Dom > > p.s. Why oh why didn't I spend my £815 UK Pounds on upgrades > for my PC's rather than on this damn PPC accellerator.. Argh! > I should have just gone bare 060, it'd much less hassle (for > all the PPC programs I use these days I might just as well). Because upgrading the pc wasn't worth it? :) -- // Jesper Svennevid : jesper@fxrealm.com Software Developer at FXRealm Studios http://www.fxrealm.com/ ------------------------------------------------------------------------ DON'T HATE YOUR RATE! Get a NextCard Visa, in 30 seconds! Get rates as low as 0.0% Intro or 9.9% Fixed APR and no hidden fees. Apply NOW! http://click.egroups.com/1/2120/0/_/451227/_/953227834/ ------------------------------------------------------------------------ @ENDNODE @NODE "Mail_210" " Re: How to find the Current Window? " From: Colin Wenzel Date: 03-16-00 15:57:33 Subject: @{b}Re: How to find the Current Window?@{ub} From: Colin Wenzel On 16-Mar-00, David Erman wrote: > From: David Erman > Hiya! > > I was wondering, If I've got this background process running, > which isn't attached to a screen/window in a natural way, how do I > find out the current window I need for attaching my EasyRequest()'s > to? > / Dave Forbid(); windowpointer = NULL; if( Process = FindTask( processname )) windowpointer = Process->pr_WindowPtr; Permit(); // test 'windowpointer' for non NULL condition here... // however, 'windowpointer' can become invalid // instantly after the Permit() if the process is exiting... // so make sure your process cannot exit before you // you want it to.... // PS: remember that -1L is also a valid value for // pr_WindowPtr if you want no requesters poping up. ------------------------------------- Colin Wenzel. Australia. EMAIL: mailto:colstv@hotkey.net.au URL: http://www.hotkey.net.au/~colstv/ ICQ: 17608330 AMIGA: 4000T, 060/50, 150Mb RAM, O.S 3.5 --------------------------------------- ------------------------------------------------------------------------ GET A NEXTCARD VISA, in 30 seconds! Get rates as low as 0.0% Intro or 9.9% Fixed APR and no hidden fees. Apply NOW! http://click.egroups.com/1/937/0/_/451227/_/953218806/ ------------------------------------------------------------------------ @ENDNODE @NODE "Mail_211" " Re: What compiler.... (Specificly, what IDE...) " From: Tim Hanson Date: 03-16-00 19:37:07 Subject: @{b}Re: What compiler.... (Specificly, what IDE...)@{ub} From: Tim Hanson > I'm partially acronymically disabled..... I`m a PAD sufferer as well. -- Tim Hanson, by the River,Canterbury, Kent. Amiga A1200T,`040/25,603e/160,32MB Fast,BVisionPPC, 8.4GB IDE HD,ZIP SCSI,RENO x2CD SCSI,Pace 56K Don`t waste your time by visiting http://www.sinister67.freeserve.co.uk ------------------------------------------------------------------------ GET A NEXTCARD VISA, in 30 seconds! Get rates as low as 0.0% Intro or 9.9% Fixed APR and no hidden fees. Apply NOW! http://click.egroups.com/1/937/0/_/451227/_/953236842/ ------------------------------------------------------------------------ @ENDNODE @NODE "Mail_212" " Re: Debugging macro trick " From: Tim Hanson Date: 03-16-00 19:42:40 Subject: @{b}Re: Debugging macro trick@{ub} From: Tim Hanson Hi Volker Volker Barthelmann wrote this on 16-Mar-00 at 05:09:21 > From: Volker Barthelmann > > > > On Wed, 15 Mar 2000, Tim Hanson wrote: > >> For debugging purposes I`d like every call to it to also pass the functions >> name as a char *. I thought a marco like this would do the job: >> >> #define pushFunction(a) pushFunction(a,"a") > #define str2(x) #x > #define str(x) str2(x) > #define pushFunction(a) pushFunction(a,str(a)) Cheers for that. I expected someone to say RTFM. I need a C tutorial book. -- Tim Hanson, by the River,Canterbury, Kent. Amiga A1200T,`040/25,603e/160,32MB Fast,BVisionPPC, 8.4GB IDE HD,ZIP SCSI,RENO x2CD SCSI,Pace 56K Don`t waste your time by visiting http://www.sinister67.freeserve.co.uk ------------------------------------------------------------------------ MAXIMIZE YOUR CARD, MINIMIZE YOUR RATE! Get a NextCard Visa, in 30 seconds! Get rates as low as 0.0% Intro or 9.9% Fixed APR and no hidden fees. Apply NOW! http://click.egroups.com/1/2122/0/_/451227/_/953236842/ ------------------------------------------------------------------------ @ENDNODE @NODE "Mail_213" " Re: What PPC compiler (was: What compiler....) " From: Jarmo Laakkonen Date: 03-16-00 21:03:44 Subject: @{b}Re: What PPC compiler (was: What compiler....)@{ub} From: Jarmo Laakkonen Jesper wrote something like this: > homepage. (Do NOT use paladins includes unmodified when developing > warpos, you get nasty instability problems due to a few small > flaws in them.) May I ask what these modifications are? And do they apply for gcc 2.95.1 as well? -- http://www.kolumbus.fi/jami.laakkonen/ppc/index.html http://www.freespeech.org/cliffbarnes/index.html A1200 040-25MHz & PPC 603e-240MHz 2+48MB "How the fuck can my skateboard be broken. I just don't understand." -- Satanic Surfers: Don't know what to do ------------------------------------------------------------------------ DON'T HATE YOUR RATE! Get a NextCard Visa, in 30 seconds! Get rates as low as 0.0% Intro or 9.9% Fixed APR and no hidden fees. Apply NOW! http://click.egroups.com/1/2120/0/_/451227/_/953237167/ ------------------------------------------------------------------------ @ENDNODE @NODE "Mail_214" " Re: Re: What compiler.... " From: Shane O'Neill Date: 03-16-00 22:30:29 Subject: @{b}Re: Re: What compiler....@{ub} From: Shane O'Neill >From: Duane McDonnell >>From: Shane O'Neill >>I certainly don't think there's a problem here, technology and development >are >>dynamic >>and move on and change. Questions answered in archives a while back are not >>necessarily >>going to bring the same answers as that same question asked more recently. >>If you don't like it just ignore it, that's what I'd do. >You make it sound as if Amiga compilers often have major updates. >Good argument if that were true, but we all know it's not. Ok, yeah, the way I worded the mail may have made it sound that way. Though, slow development is still development and my reply is still valid, albeit to a lesser extent than I made it sound. Hmm .. Maybe I shouldn't write mails drunk. ;) Regards, Shane. ------------------------------------------------------------------------ DON'T HATE YOUR RATE! Get a NextCard Visa, in 30 seconds! Get rates as low as 0.0% Intro or 9.9% Fixed APR and no hidden fees. Apply NOW! http://click.egroups.com/1/2120/0/_/451227/_/953242324/ ------------------------------------------------------------------------ @ENDNODE @NODE "Mail_215" " ami dev cd probs " From: Linus McCabe Date: 03-16-00 22:39:06 Subject: @{b}ami dev cd probs@{ub} From: Linus McCabe yello, list! I recently bought the amidev cd 2.1 and I'm very happy with the purchase! However, I'm about to install the stromc that comes on the cd, to see if maybe its time to let sasc rest in peace. Anyway, due to the fact that the cdrom in my ami died a long time ago I have to access my cds through a samba filesharing over my tcp/ip network from my windows pc. Windows can show the XXX.¶ files when browsing the cd, but it cant copy them :/ so i've managed to copy/install everything, except for the .¶ files. I wonder if anyone who has the cd could get in touch with me to do a lha a ram:projects.lha amidevcd:whateverdir/ #?.¶ -r and send it to me later (get it confirmed first, so not 20+ ppl decides to send me huge emails ;) I'd appreciate it very much! /Linus -- Linus McCabe * Sparkle@hehe.com * Sparkle, #Amiga, DalNet * http://sparkle.amiga.tm You can't have the blues in an office or a shopping mall, the lighting is wrong. ------------------------------------------------------------------------ MAXIMIZE YOUR CARD, MINIMIZE YOUR RATE! Get a NextCard Visa, in 30 seconds! Get rates as low as 0.0% Intro or 9.9% Fixed APR and no hidden fees. Apply NOW! http://click.egroups.com/1/2122/1/_/451227/_/953631310/ ------------------------------------------------------------------------ @ENDNODE @NODE "Mail_216" " RE: Re: What PPC compiler (was: What compiler....) " From: Jesper Svennevid Date: 03-16-00 23:00:17 Subject: @{b}RE: Re: What PPC compiler (was: What compiler....)@{ub} From: Jesper Svennevid > From: Jarmo Laakkonen > > Jesper wrote something like this: > > > homepage. (Do NOT use paladins includes unmodified when developing > > warpos, you get nasty instability problems due to a few small > > flaws in them.) > > May I ask what these modifications are? And do they apply for > gcc 2.95.1 as well? They might. Every gcc compile for amigappc I've used except the original one by GG contain some kind of problem with alloca(), so when alloca() is called, it never resets the pointer that it allocated from. After a while, this will exceed the limit of the stack, and start tampering with other processes memory. For example, I wrote a small video-overlay test(using cgxvideo.library), and with unmodified includes it locked up after about 20-30 seconds, but when I modified the includes to use a "normal" stack-allocation (letting the function-context allocate one for you, not using alloca()), I had it running for 6 hours before I shut it off. This was fixed in the march release of powerup-includes, but the egcs- warpos wrappers (as they share the inline-files with powerup) on paladins homepage are based on an earlier version, and thus still contains the problem. -- // Jesper Svennevid : jesper@fxrealm.com Software Developer at FXRealm Studios http://www.fxrealm.com/ ------------------------------------------------------------------------ MAXIMIZE YOUR CARD, MINIMIZE YOUR RATE! Get a NextCard Visa, in 30 seconds! Get rates as low as 0.0% Intro or 9.9% Fixed APR and no hidden fees. Apply NOW! http://click.egroups.com/1/2122/0/_/451227/_/953244047/ ------------------------------------------------------------------------ @ENDNODE @NODE "Mail_217" " Re: Better editor than se in SAS/C 6.0? " From: Stephen Illingworth Date: 03-16-00 23:19:26 Subject: @{b}Re: Better editor than se in SAS/C 6.0?@{ub} From: Stephen Illingworth Hello Kelly On 16-Mar-00, you wrote: > From: Kelly Samel > > Hello David > > On 15-Mar-00, you wrote: > >> From: David Erman >> >> >> Is it just me using emacs? >> >> / Dave > > I think you are the only one left. ;) > Hehe! Personally I use VIm. I can hear the groans now, but it really is an excellent editor once you get used to it. --- Steve ------------------------------------------------------------------------ MAXIMIZE YOUR CARD, MINIMIZE YOUR RATE! Get a NextCard Visa, in 30 seconds! Get rates as low as 0.0% Intro or 9.9% Fixed APR and no hidden fees. Apply NOW! http://click.egroups.com/1/2122/0/_/451227/_/953249558/ ------------------------------------------------------------------------ @ENDNODE @NODE "Mail_218" " Re: stci_d() (was: Re: Better editor than se in SAS/C 6.0?) " From: Kelly Samel Date: 03-17-00 02:28:14 Subject: @{b}Re: stci_d() (was: Re: Better editor than se in SAS/C 6.0?)@{ub} From: Kelly Samel Hello Olaf On 16-Mar-00, you wrote: > From: "Olaf Barthel" >>> [...] Also does Storm or HiSoft C have an equivalent >>> function to stci_d()? I have been unable to find one. >> >> What is it? > > It converts a number into a string (this is a non-portable > SAS/C function). This makes you wonder why Kelly doesn't > use sprintf() instead ;) Amazing, how could I have missed that one? I will start using it straight away Mr. Barthel. ;) This solves one of the main problems when porting my Sas/C code to Storm and HiSoft, thanks. ------------------------------------------------------------------------ DON'T HATE YOUR RATE! Get a NextCard Visa, in 30 seconds! Get rates as low as 0.0% Intro or 9.9% Fixed APR and no hidden fees. Apply NOW! http://click.egroups.com/1/2120/0/_/451227/_/953257205/ ------------------------------------------------------------------------ @ENDNODE @NODE "Mail_219" " Re: Re: What compiler.... " From: Volker Barthelmann Date: 03-17-00 02:34:30 Subject: @{b}Re: Re: What compiler....@{ub} From: Volker Barthelmann On Thu, 16 Mar 2000, Hynek Schlawack wrote: > > Apart from putting frontend, preprocessor, compiler, assembler and linker > > into RAM, no. Actually, I think vbcc compiles about as fast as most other > > they ain't pure, richt? Right. But loading the compiler should be barely noticeable unless you have a really slow system. > >> Now I remebered a > >> third problem..."make" is crashing after linking (better: the console > >> handler)!? Has anyone this problem too? DMake doesn't have this > >> problem... > > Perhaps the stack size is too small or your make has problems? > > I don't think, that the stack is the problem as I have a stack of 124.000 > bytes. And I'm talking about the GNU-make which hasn't crashed before even > while using a stack-usage-moster like gcc :-). If you can reproduce the problem, send me some details, but it really doesn't sound like a problem of vbcc. Volker ------------------------------------------------------------------------ DON'T HATE YOUR RATE! Get a NextCard Visa, in 30 seconds! Get rates as low as 0.0% Intro or 9.9% Fixed APR and no hidden fees. Apply NOW! http://click.egroups.com/1/2120/0/_/451227/_/953257014/ ------------------------------------------------------------------------ @ENDNODE @NODE "Mail_220" " Re: What PPC compiler (was: What compiler....) " From: Gavin Kinsey Date: 03-17-00 08:05:56 Subject: @{b}Re: What PPC compiler (was: What compiler....)@{ub} From: Gavin Kinsey I received a message from Dominic Clifton, a reply follows, > I managed to get a few small programs working with VBCC, but the compiler > crashes randomly and so did the compiled executables, and the PPC versions > of my programs turned out to be slower than the native 680x0 versions. The speed of the code depends very much on your design, if you are doing OS calls every other line then it will be slower. > EGCS was way to much of a pain in the ass to set up, and it wouldn't cope > with most of my SAS/C code, I needed to create a great big list of > #define's to stop it complaining and I didn't have the time (has anyone > got any include files which make it easier to port SAS/C code to other > compilers ? I'd be *eternally* grateful!) gcc is much easier to set up now, all you have to do is download the archive from my site and unlzx to GG:. If you have gcc working for 68k then that should be enough to get it working for ppc, it might work on its own too but I'm not sure about that. >> vbcc seems to work quite well, it does at least have working asin/acos >> functions. > > Hehe, yeah, pity it crashes on my 604/233 to damn much though. Are you sure you are not experiencing the hardware flaw? I think it's called the LR bug, it only seems to affect 233MHz boards. Do other ppc programs crash at random, try running rc5des_ppc for a few hours, if that crashes you may have the bug. -- Man invented language to satisfy his deep need to complain. - Lily Tomlin Gavin Kinsey ------------------------------------------------------------------------ MAXIMIZE YOUR CARD, MINIMIZE YOUR RATE! Get a NextCard Visa, in 30 seconds! Get rates as low as 0.0% Intro or 9.9% Fixed APR and no hidden fees. Apply NOW! http://click.egroups.com/1/2122/1/_/451227/_/953340285/ ------------------------------------------------------------------------ @ENDNODE @NODE "Mail_221" " GadTools listviews " From: andrewmarkwell@ukonline.co.uk Date: 03-17-00 10:39:02 Subject: @{b}GadTools listviews@{ub} From: andrewmarkwell@ukonline.co.uk Hi, I've got a problem with trying to create a GadTools listview, the listview displays, but it won't show any of the text inside the gadget: I have attached my code, could anyone tell me what's wrong with it? (apart from not freeing up resources!) void main(void) { Window *window; NewGadget ng; TextAttr ta={"helvetica.font", 15, FS_NORMAL, FPF_DISKFONT}; APTR vi; Screen *screen; Gadget *glist=NULL, *gad; List mylist; Node *node; IntuitionBase=OpenLibrary("intuition.library", 39); GadToolsBase=OpenLibrary("gadtools.library", 39); DiskfontBase=OpenLibrary("diskfont.library", 39); NewList(&mylist); node=new Node; node->ln_Name="item1"; AddTail(&mylist, node); node=new Node; node->ln_Name="item2"; AddTail(&mylist, node); node=new Node; node->ln_Name="item3"; AddTail(&mylist, node); screen=LockPubScreen(NULL); OpenDiskFont(&ta); vi=GetVisualInfo(screen, TAG_END); gad=CreateContext(&glist); window=OpenWindowTags(NULL, WA_Left, 50, WA_Top, 50, WA_Width, 400, WA_Height, 200, TAG_END); ng.ng_LeftEdge=10; ng.ng_TopEdge=15; ng.ng_Width=300; ng.ng_Height=150; ng.ng_GadgetText="Listview"; ng.ng_TextAttr=&ta; ng.ng_GadgetID=5; ng.ng_Flags=PLACETEXT_ABOVE; ng.ng_VisualInfo=vi; ng.ng_UserData=NULL; gad=CreateGadget(LISTVIEW_KIND, gad, &ng, GTLV_Labels, (List*)&mylist, GTLV_ReadOnly, TRUE, TAG_END); AddGList(window, glist, -1, -1, NULL); RefreshGList(glist, window, NULL, -1); system("C:Wait 10"); RemoveGList(window, glist, -1); FreeGadgets(glist); CloseWindow(window); } ------------------------------------------------------------------------ MAXIMIZE YOUR CARD, MINIMIZE YOUR RATE! Get a NextCard Visa, in 30 seconds! Get rates as low as 0.0% Intro or 9.9% Fixed APR and no hidden fees. Apply NOW! http://click.egroups.com/1/2122/1/_/451227/_/953285944/ ------------------------------------------------------------------------ @ENDNODE @NODE "Mail_222" " Re: How to find the Current Window? " From: David Erman Date: 03-17-00 12:15:43 Subject: @{b}Re: How to find the Current Window?@{ub} From: David Erman Hi Duane, In a mail of 18-Mar-00, you wrote: DM>> It's not valid unless the device task is a process (most aren't). DM>> If the code froze you're almost certainly a basic task so the write DM>> to pr_WindowPtr went to memory you don't own (since there's no Process DM>> structure extension). DM> That should be *read* from pr_WindowPtr came from invalid memory, DM> hence you get junk for the window pointer and a subsequent hang, DM> crash, fireworks, whatever when calling EasyRequest(). Actually, this was a plain let's-compile-the-examples-from-SAS\C-and-see-what-happens thing, so the process is started with myProc = CreateNewProcTags(NP_Entry, cmd_handler, NP_StackSize, 4096, NP_Name, "porno device test", TAG_DONE); Shouldn't that make it a Process? Yours, David -- Studying Computer Sciences & Signal Processing UIN: 5361039 A4000/Apollo 040@40MHz/32Mb/1.6Gb/4X Mitsumi/M1438 RedHat Linux 6.0 -- AmigaOS 3.1 -- MacOS 8.1 ------------------------------------------------------------------------ PERFORM CPR ON YOUR APR! Get a NextCard Visa, in 30 seconds! Get rates as low as 0.0% Intro or 9.9% Fixed APR and no hidden fees. Apply NOW! http://click.egroups.com/1/2121/1/_/451227/_/953291565/ ------------------------------------------------------------------------ @ENDNODE @NODE "Mail_223" " Re: Better editor than se in SAS/C 6.0? " From: Kelly Samel Date: 03-17-00 12:17:53 Subject: @{b}Re: Better editor than se in SAS/C 6.0?@{ub} From: Kelly Samel Hello David On 17-Mar-00, you wrote: > From: David Erman > In a mail of 16-Mar-00, you wrote: > > >> Hehe! Personally I use VIm. I can hear the groans now, but it really >> excellent editor once you get used to it. > > Yeah, but no syntax highliting on the Amiga, right? I like syntax > highlighting. > > Yours, > David ALL of the Amiga compilers with a IDE have syntax highlighting. Also GoldEd and a few others do too. In fact the syntax highlighting on GoldEd goes beyond most programs in the configurability department. If you mean that Vim in particular does not have the highlighting then you could be right but there are others available and I am sure the latest Vim could be ported pretty easily. ------------------------------------------------------------------------ GET A NEXTCARD VISA, in 30 seconds! Get rates as low as 0.0% Intro or 9.9% Fixed APR and no hidden fees. Apply NOW! http://click.egroups.com/1/937/1/_/451227/_/953292814/ ------------------------------------------------------------------------ @ENDNODE @NODE "Mail_224" " Re: How to find the Current Window? " From: David Erman Date: 03-17-00 12:53:20 Subject: @{b}Re: How to find the Current Window?@{ub} From: David Erman Hi Duane, In a mail of 18-Mar-00, you wrote: DM>> Shouldn't that make it a Process? DM> It's impossible to say without seeing the DM> full code. It's basically the example.device that comes with SAS\C. I'm mucking about with the code, just to get aqcuainted with the concepts of writing a device driver. DM> You gotta be careful when playing with pr_WindowPtr, if you set DM> it to a window pointer the window obviously needs to be valid DM> for the duration of the setting (easy to forget this). Also, DM> always set it back to the original value when you exit a task. DM> This is absolutely required for undetached shell startups and DM> a good habit to get into for all invocation methods in general. I used the pointer when closing the device. Not a good idea, then? Whaddya mean by setting it back to the original value? EasyRequest() doesn't change the windowpointer, does it? Yours, David -- Studying Computer Sciences & Signal Processing UIN: 5361039 A4000/Apollo 040@40MHz/32Mb/1.6Gb/4X Mitsumi/M1438 RedHat Linux 6.0 -- AmigaOS 3.1 -- MacOS 8.1 ------------------------------------------------------------------------ GET A NEXTCARD VISA, in 30 seconds! Get rates as low as 2.9% Intro or 9.9% Fixed APR and no hidden fees. Apply NOW! http://click.egroups.com/1/936/1/_/451227/_/953293826/ ------------------------------------------------------------------------ @ENDNODE @NODE "Mail_225" " Re: Better editor than se in SAS/C 6.0? " From: David Erman Date: 03-17-00 13:05:38 Subject: @{b}Re: Better editor than se in SAS/C 6.0?@{ub} From: David Erman Hi Kelly, In a mail of 17-Mar-00, you wrote: KS> ALL of the Amiga compilers with a IDE have syntax highlighting. KS> Also GoldEd and a few others do too. In fact the syntax KS> highlighting on GoldEd goes beyond most programs in the KS> configurability department. If you mean that Vim in particular KS> does not have the highlighting then you could be right but KS> there are others available and I am sure the latest Vim could KS> be ported pretty easily. I find GoldEd big and bulky, StormC buggy and odd. emacs, however, is big, bulky, odd, and quite usable. :) There is a port of vim already, but without syntaxcolouring, it only runs in a console window. Not very pretty to look at, and veruy obscure to lear if you're used to emacs. Yours, David -- Studying Computer Sciences & Signal Processing UIN: 5361039 A4000/Apollo 040@40MHz/32Mb/1.6Gb/4X Mitsumi/M1438 RedHat Linux 6.0 -- AmigaOS 3.1 -- MacOS 8.1 ------------------------------------------------------------------------ DON'T HATE YOUR RATE! Get a NextCard Visa, in 30 seconds! Get rates as low as 0.0% Intro or 9.9% Fixed APR and no hidden fees. Apply NOW! http://click.egroups.com/1/2120/1/_/451227/_/953294560/ ------------------------------------------------------------------------ @ENDNODE @NODE "Mail_226" " Re: Re: What compiler.... " From: Duane McDonnell Date: 03-17-00 14:54:49 Subject: @{b}Re: Re: What compiler....@{ub} From: Duane McDonnell >From: Shane O'Neill >I certainly don't think there's a problem here, technology and development are >dynamic >and move on and change. Questions answered in archives a while back are not >necessarily >going to bring the same answers as that same question asked more recently. >If you don't like it just ignore it, that's what I'd do. You make it sound as if Amiga compilers often have major updates. Good argument if that were true, but we all know it's not. ------------------------------------------------------------------------ PERFORM CPR ON YOUR APR! Get a NextCard Visa, in 30 seconds! Get rates as low as 0.0% Intro or 9.9% Fixed APR and no hidden fees. Apply NOW! http://click.egroups.com/1/2121/0/_/451227/_/953238516/ ------------------------------------------------------------------------ @ENDNODE @NODE "Mail_227" " Inline " From: Bart King Date: 03-17-00 17:08:07 Subject: @{b}Inline@{ub} From: Bart King Hi. Something that's been bugging me for ages is whether or not it is possible to use inline asm code in C code on Amiga compilers? It seems to be a waste of time optimising small parts of C code in asm due to the overheads gained from having three lines of asm in a separate object. Like MSVC: for(int blah = 0; blah < 20; blah++) { _asm { ... } } I know EGCS supports inline asm from Linux, but unsure about StormC (manual is really useless), SAS/C, etc... -- Bart King Developer http://www.bartman.demon.co.uk (not up yet - soon) http://www.fscked.net/fubar/ ICQ: 18178781 ------------------------------------------------------------------------ PERFORM CPR ON YOUR APR! Get a NextCard Visa, in 30 seconds! Get rates as low as 0.0% Intro or 9.9% Fixed APR and no hidden fees. Apply NOW! http://click.egroups.com/1/2121/1/_/451227/_/953332568/ ------------------------------------------------------------------------ @ENDNODE @NODE "Mail_228" " BGUI WindowOpen " From: Shane O'Neill Date: 03-17-00 17:21:23 Subject: @{b}BGUI WindowOpen@{ub} From: Shane O'Neill Hi, Hmmm, I can't seem to figure out how to position Windows using the BGUI WindowOpen macro, probably due to poor documentation. Anyway, there's a tag called WINDOW_Position, though the only values that this accepts are POS_CENTERWINDOW, POS_CENTERMOUSE and POS_TOPLEFT. I really need to be more specific. If I have to I can use OpenWindowTags() and some other functions, though I'd prefer to get it working with WindowOpen. Anyone know how to do this? Regards, Shane _____________________________ There is a place that still remains It eats the fear it eats the pain -NIN _____________________________ ------------------------------------------------------------------------ DON'T HATE YOUR RATE! Get a NextCard Visa, in 30 seconds! Get rates as low as 0.0% Intro or 9.9% Fixed APR and no hidden fees. Apply NOW! http://click.egroups.com/1/2120/1/_/451227/_/953310345/ ------------------------------------------------------------------------ @ENDNODE @NODE "Mail_229" " Re: How to find the Current Window? " From: David Erman Date: 03-17-00 18:49:12 Subject: @{b}Re: How to find the Current Window?@{ub} From: David Erman Hi Duane, In a mail of 18-Mar-00, you wrote: DM> I had a quick look at the code you're referring to but I can't see why DM> that would hang. You're welcome to mail me the modified code if you DM> want a proper explanation, but it sounds like the problem is solved DM> now so it's all rather academic :-) I agree. Let's cut it short here, before this thread deteriorates to some pointless off-topic thing. Thaks for the help tho'! Yours, David -- Studying Computer Sciences & Signal Processing UIN: 5361039 A4000/Apollo 040@40MHz/32Mb/1.6Gb/4X Mitsumi/M1438 RedHat Linux 6.0 -- AmigaOS 3.1 -- MacOS 8.1 ------------------------------------------------------------------------ DON'T HATE YOUR RATE! Get a NextCard Visa, in 30 seconds! Get rates as low as 0.0% Intro or 9.9% Fixed APR and no hidden fees. Apply NOW! http://click.egroups.com/1/2120/1/_/451227/_/953315178/ ------------------------------------------------------------------------ @ENDNODE @NODE "Mail_230" " Re: Inline " From: Ilkka Lehtoranta Date: 03-18-00 00:05:05 Subject: @{b}Re: Inline@{ub} From: Ilkka Lehtoranta On 17-Maa-00, Bart King wrote: BK> Something that's been bugging me for ages is whether or not it is BK> possible to use inline asm code in C code on Amiga compilers? It seems BK> to be a waste of time optimising small parts of C code in asm due to the BK> overheads gained from having three lines of asm in a separate object. I don't know why you want inline asm code. Very likely users can't see any difference in speed. Not even if they are using a vanilla A1200. Optimizing C code in asm _is_ a waste of time. ------------------------------------------------------------------------ GET A NEXTCARD VISA, in 30 seconds! Get rates as low as 0.0% Intro APR and no hidden fees. Apply NOW! http://click.egroups.com/1/975/1/_/451227/_/953334724/ ------------------------------------------------------------------------ @ENDNODE @NODE "Mail_231" " What compiler / MakeFile " From: "Alastair M. Robinson" Date: 03-18-00 01:08:29 Subject: @{b}What compiler / MakeFile@{ub} From: "Alastair M. Robinson" Hi Colin, > Enclosed is a program I use to check the archive bit to > see if any of the modules have changed... > When I compile them to an object file, I protect the > source file with +A Great minds think alike - I've been using this method ever since I started using VBCC! The reason I don't use make is that my A4000/030's clock battery doesn't work any more. (It's a Lithium Coin, not a Ni-Cad, so I'm not too worried about it leaking, and I'm not brave enough to replace it yet!) I suppose what would be useful is a program to detect when a header file has changed, and touch the `A' flag of all files which depend on it. My contribution to the ongoing discussion: I use VBCC because it's free, produces good code, is easy to set up, and with care it can be used to compile 68k, PPC-PowerUP and WarpOS versions of a project from a single source tree. (Though the 'A' trick doesn't work too well in that case!) -- Alastair M. Robinson, email:blackfive@fakenhamweb.co.uk Amiga 4000/030, 18Meg RAM, 4.3Gig HD, Mustek ScanExpress 6000SP C, Assembler (68k, PIC16C84, ATMEL AVR), PostScript, ARexx, AMOSPro(!), AmigaDOS Experience is directly proportional to the value of equipment destroyed. - Carolyn Scheppner ------------------------------------------------------------------------ GET A NEXTCARD VISA, in 30 seconds! Get rates as low as 0.0% Intro or 9.9% Fixed APR and no hidden fees. Apply NOW! http://click.egroups.com/1/937/1/_/451227/_/953338254/ ------------------------------------------------------------------------ @ENDNODE @NODE "Mail_232" " Re: Re: How to find the Current Window? " From: Duane McDonnell Date: 03-18-00 06:06:14 Subject: @{b}Re: Re: How to find the Current Window?@{ub} From: Duane McDonnell >From: David Erman > Hi Duane, >In a mail of 18-Mar-00, you wrote: >DM>> It's not valid unless the device task is a process (most aren't). >DM>> If the code froze you're almost certainly a basic task so the write >DM>> to pr_WindowPtr went to memory you don't own (since there's no Process >DM>> structure extension). >DM> That should be *read* from pr_WindowPtr came from invalid memory, >DM> hence you get junk for the window pointer and a subsequent hang, >DM> crash, fireworks, whatever when calling EasyRequest(). > Actually, this was a plain >let's-compile-the-examples-from-SAS\C-and-see-what-happens thing, so the >process is started with > myProc = CreateNewProcTags(NP_Entry, cmd_handler, > NP_StackSize, 4096, > NP_Name, "porno device test", > TAG_DONE); > Shouldn't that make it a Process? Yes, it does. That makes your hang rather weird. Perhaps you were getting -1 for the window pointer and not handling that properly, or maybe pr_WindowPtr was no longer valid since the window was subsequently closed? It's impossible to say without seeing the full code. You gotta be careful when playing with pr_WindowPtr, if you set it to a window pointer the window obviously needs to be valid for the duration of the setting (easy to forget this). Also, always set it back to the original value when you exit a task. This is absolutely required for undetached shell startups and a good habit to get into for all invocation methods in general. ------------------------------------------------------------------------ PERFORM CPR ON YOUR APR! Get a NextCard Visa, in 30 seconds! Get rates as low as 0.0% Intro or 9.9% Fixed APR and no hidden fees. Apply NOW! http://click.egroups.com/1/2121/1/_/451227/_/953293256/ ------------------------------------------------------------------------ @ENDNODE @NODE "Mail_233" " Re: ami dev cd probs " From: Linus McCabe Date: 03-18-00 07:08:00 Subject: @{b}Re: ami dev cd probs@{ub} From: Linus McCabe On 21-Mar-00, Matthew Fletcher wrote: > Easy just zip them,... or tgz or lha, or zip the whole stormc dir > and then remove all but the .¶ files. i tried ot zip them using winzip, but the pc cant read the .¶ files at all (complains that the files are not found, althought they're seen in the list) But Dirk Harklaar has promised me to send me the files, so the problem is solved. Thanks Dirk, and everyone else who had suggestions and offers! /Linus > > -matthew > > > > > ------------------------------------------------------------------------ > PERFORM CPR ON YOUR APR! > Get a NextCard Visa, in 30 seconds! Get rates as low as 0.0% Intro > or 9.9% Fixed APR and no hidden fees. Apply NOW! > http://click.egroups.com/1/2121/1/_/451227/_/953648105/ > ------------------------------------------------------------------------ > > > -- Linus McCabe * Sparkle@hehe.com * Sparkle, #Amiga, DalNet * http://sparkle.amiga.tm Is there a new world rising, from the shadows of the old? ------------------------------------------------------------------------ GET A NEXTCARD VISA, in 30 seconds! Get rates as low as 0.0% Intro APR and no hidden fees. Apply NOW! http://click.egroups.com/1/975/1/_/451227/_/953659984/ ------------------------------------------------------------------------ @ENDNODE @NODE "Mail_234" " Re: Re: How to find the Current Window? " From: Duane McDonnell Date: 03-18-00 07:52:53 Subject: @{b}Re: Re: How to find the Current Window?@{ub} From: Duane McDonnell >From: David Erman > It's basically the example.device that comes with SAS\C. I'm mucking about >with the code, just to get aqcuainted with the concepts of writing a >device driver. I had a quick look at the code you're referring to but I can't see why that would hang. You're welcome to mail me the modified code if you want a proper explanation, but it sounds like the problem is solved now so it's all rather academic :-) >DM> You gotta be careful when playing with pr_WindowPtr, if you set >DM> it to a window pointer the window obviously needs to be valid >DM> for the duration of the setting (easy to forget this). Also, >DM> always set it back to the original value when you exit a task. >DM> This is absolutely required for undetached shell startups and >DM> a good habit to get into for all invocation methods in general. > I used the pointer when closing the device. Not a good idea, then? Whaddya >mean by setting it back to the original value? I mean, restore it to its original value prior to exit if you changed it yourself (you can set the pointer as well as read it). If you don't change it, there's no need to do this. > EasyRequest() doesn't change the windowpointer, does it? No, it would be illegal for function calls to alter it (they could legally change it for their own purposes, but they'd have to restore it before returning to the calling function; I can't think of an occasion where this would be required though). ------------------------------------------------------------------------ GET A NEXTCARD VISA, in 30 seconds! Get rates as low as 0.0% Intro APR and no hidden fees. Apply NOW! http://click.egroups.com/1/975/1/_/451227/_/953299800/ ------------------------------------------------------------------------ @ENDNODE @NODE "Mail_235" " Re: Re: Inline " From: Olaf Barthel Date: 03-18-00 10:43:16 Subject: @{b}Re: Re: Inline@{ub} From: "Olaf Barthel" On Mar 18 Ilkka (Ilkka Lehtoranta) wrote: > From: Ilkka Lehtoranta > > On 17-Maa-00, Bart King wrote: > > BK> Something that's been bugging me for ages is whether or not it is > BK> possible to use inline asm code in C code on Amiga compilers? It seems > BK> to be a waste of time optimising small parts of C code in asm due to the > BK> overheads gained from having three lines of asm in a separate object. > > I don't know why you want inline asm code. Very likely users can't see any > difference in speed. Not even if they are using a vanilla A1200. > > Optimizing C code in asm _is_ a waste of time. I disagree. It can make sense, but it is rarely necessary today. Most `C' compilers we use today will use CPU registers much more efficiently than the compilers available 10-12 years ago (and optimizations performed by "current" compilers are of course a whole magnitude better). And for some rare applications there can be a noticeable speed improvement if the code translated into assembly language fits into the CPU cache. I once spent a week tweaking a Huffman decoder to fit (well, at least the important bits) into the cache of the '030. Thankfully, such acrobatics are no longer necessary today :) -- Home: Olaf Barthel, Brabeckstrasse 35, D-30559 Hannover Net: olsen@sourcery.han.de (Home), olsen@logicalline.com (Work) ------------------------------------------------------------------------ GET A NEXTCARD VISA, in 30 seconds! Get rates as low as 0.0% Intro APR and no hidden fees. Apply NOW! http://click.egroups.com/1/975/1/_/451227/_/953405919/ ------------------------------------------------------------------------ @ENDNODE @NODE "Mail_236" " Opening libraries... " From: Alexander Niven-Jenkins Date: 03-18-00 11:42:04 Subject: @{b}Opening libraries...@{ub} From: Alexander Niven-Jenkins Hiya, What is wrong with this? I have used this sequence to open a library in everything even dating back to NorthC but with StormC it chucks... struct Library *UtilityBase = NULL; if( UtilityBase = OpenLibrary( UTILITYNAME, 39 ) ) { // code here if( UtilityBase ) { CloseLibrary( UtilityBase ); } } Same goes for LocaleBase? However MUIMasterBase works? BTW the errors are: Syntax error in expression or declarator and Assignment: Left value must be L-value, these are where ever UtilityBase is used? Help! All this just to get old code to compile so that I can finally scrap HiSOFT! Kind regards... Alex -- Cremlin Software - http://www.cremlinsoftware.org Founder, lead programmer - ghandi@cremlinsoftware.org Production, programmer - anj@cremlinsoftware.org ------------------------------------------------------------------------ GET A NEXTCARD VISA, in 30 seconds! Get rates as low as 0.0% Intro APR and no hidden fees. Apply NOW! http://click.egroups.com/1/975/1/_/451227/_/953376173/ ------------------------------------------------------------------------ @ENDNODE @NODE "Mail_237" " RE: Opening libraries... " From: andrewmarkwell@ukonline.co.uk Date: 03-18-00 11:51:08 Subject: @{b}RE: Opening libraries...@{ub} From: andrewmarkwell@ukonline.co.uk On Sat, 18 Mar 2000 10:42:04 +0000, Alexander Niven-Jenkins wrote: > From: Alexander Niven-Jenkins > > Hiya, > > What is wrong with this? I have used this sequence to open a libr > ary in > everything even dating back to NorthC but with StormC it chucks... > > struct Library *UtilityBase = NULL; > > if( UtilityBase = OpenLibrary( UTILITYNAME, 39 ) ) > { > // code here > if( UtilityBase ) > { > CloseLibrary( UtilityBase ); > } > } > > Same goes for LocaleBase? However MUIMasterBase works? What is UTILITYNAME? Does it work if you replace it with "utility.library"? > Help! All this just to get old code to compile so that I can fina > lly > scrap HiSOFT! HiSoft rocks! Andrew Markwell ------------------------------------------------------------------------ PERFORM CPR ON YOUR APR! Get a NextCard Visa, in 30 seconds! Get rates as low as 0.0% Intro or 9.9% Fixed APR and no hidden fees. Apply NOW! http://click.egroups.com/1/2121/1/_/451227/_/953376669/ ------------------------------------------------------------------------ @ENDNODE @NODE "Mail_238" " Re: Re: Inline " From: Bart King Date: 03-18-00 12:24:31 Subject: @{b}Re: Re: Inline@{ub} From: Bart King Ilkka Lehtoranta wrote: > I don't know why you want inline asm code. Very likely users can't see any > difference in speed. Not even if they are using a vanilla A1200. Well, I'm trying to optimize my chunky graphics system, because it seems to take an ice age (like a second or two) to copy a 640x480 chunky buffer onto another 640x480 buffer, whilst skipping over colour 255. I suppose the ideal situation would be to re-write the entire system (or at least the time-consuming bits) in asm. But, I'm not really a major league asm programmer (I know enough to get by, but not everything). > Optimizing C code in asm _is_ a waste of time. Heh yes, but replacing it (or parts of it) with optimized asm code isn't :) -- Bart King Developer http://www.bartman.demon.co.uk (not up yet - soon) http://www.fscked.net/fubar/ ICQ: 18178781 ------------------------------------------------------------------------ DON'T HATE YOUR RATE! Get a NextCard Visa, in 30 seconds! Get rates as low as 0.0% Intro or 9.9% Fixed APR and no hidden fees. Apply NOW! http://click.egroups.com/1/2120/1/_/451227/_/953403951/ ------------------------------------------------------------------------ @ENDNODE @NODE "Mail_239" " Re: Opening libraries... " From: Alexander Niven-Jenkins Date: 03-18-00 12:27:40 Subject: @{b}Re: Opening libraries...@{ub} From: Alexander Niven-Jenkins Hiya andrewmarkwell@ukonline.co.uk, On 18-Mar-00, you wrote: > What is UTILITYNAME? Does it work if you replace it > with "utility.library"? It's the libraries name as defined within the includes: utility/utility.h #define UTILITYNAME "utility.library" Nope this isn't he problem, it's moaning about UtilityBase. >> Help! All this just to get old code to compile so that I can >> finally scrap HiSOFT! > HiSoft rocks! Apart from the fact that the IDE isn't too hot with RTG and it's dead :-) Other than that I liked it, I found it easier to get into than Storm, although after 3 years I've decided to migrate to an IDE based compiler that is still in development. Especially as it came on my developer CD :-) Kind regards... Alex -- Cremlin Software - http://www.cremlinsoftware.org Founder, lead programmer - ghandi@cremlinsoftware.org Production, programmer - anj@cremlinsoftware.org ------------------------------------------------------------------------ PERFORM CPR ON YOUR APR! Get a NextCard Visa, in 30 seconds! Get rates as low as 0.0% Intro or 9.9% Fixed APR and no hidden fees. Apply NOW! http://click.egroups.com/1/2121/1/_/451227/_/953378894/ ------------------------------------------------------------------------ @ENDNODE @NODE "Mail_240" " Re: Opening libraries... " From: Hynek Schlawack Date: 03-18-00 14:56:06 Subject: @{b}Re: Opening libraries...@{ub} From: Hynek Schlawack Am 18-Mär-00 schrieb Alexander Niven-Jenkins: >> What is UTILITYNAME? Does it work if you replace it >> with "utility.library"? > It's the libraries name as defined within the includes: > utility/utility.h > #define UTILITYNAME "utility.library" > Nope this isn't he problem, it's moaning about UtilityBase. Try to not define the UtilityBase. Gruss, -- Hynek Schlawack ++ 7Wire Technologies ++ http://me.in-berlin.de/~hys ------------------------------------------------------------------------ GET A NEXTCARD VISA, in 30 seconds! Get rates as low as 0.0% Intro APR and no hidden fees. Apply NOW! http://click.egroups.com/1/975/1/_/451227/_/953387759/ ------------------------------------------------------------------------ @ENDNODE @NODE "Mail_241" " Re: What compiler.... (Specificly, what IDE...) " From: Gabriele Svelto Date: 03-18-00 15:37:51 Subject: @{b}Re: What compiler.... (Specificly, what IDE...)@{ub} From: Gabriele Svelto Hi Dominic > From: "Dominic Clifton" > > Here's my 2 cents on this... > > After using recent IDE's on PC system's I'm amazed that I coped for so > long with the tools that I've had on my Amigas. Even borland's DOS > based Turbo Pascal/Turbo C had better IDE's than *any* Amiga IDE has > ever had. > > These days IDE's on PC's have built in GUI editors, tab-completion of > function names, excellent large project management, function lookup, > on-line context sensitive help (of the IDE and the > C-functions/keywords/APIs etc) they're just downright brilliant to > work in, everything's at your fingertips. > > Granted, they may not be as configurable as an Amiga (no other > computer system has, or will be in my opinion) but at least the > essentials for development are right there. > > If I had the tools on my Amiga that the PC world enjoys today I'm sure > I'd still be coding on it as much as I used to, and I'd be able to > spend more time coding than trying to get the damn development > programs working! I think you're missing one point: there isn't an absolutely better IDE on Amiga or PeeCees, the best IDE is the one you use it better and are more productive with. I couldn't give up shells, I'm enormously more productive with a highly configurable, self-made combination of tools running from a shell than on a rigid all-in-one IDE, on the other hand somebody could prefer an all in one proggie but this is matter of taste. > I've tried every amiga compiler that I've found (North-C, Manx, SAS/C, > Storm, Hisoft, GCC, VBCC) and I keep trying new versions of each one, > and I've never come across anything as easy to develop in than as the > old DOS based Turbo C was. The nicest IDE I found on the amiga was the > new Hisoft-C, but the compiler wouldn't compile most of my SAS/C > specific code so I gave up on it as I didn't have the time. The main > problem with Amiga IDE's (and compilers), after lack of essential > features named above, is the fact that they're either not properly > documented, or if they are the documentation is in another language > (Mainly german). Yes, Amiga compilers probably lack some features (but nont really much) but finding developement tools on the Amiga is a lot easier than on Windows or other commercial OSes (such as MacOS). Take for example the DevCD, 50 DM, everything to start developping, Microsoft developer CDs cost over 1800DM (at least in my country)! And if you don't like Amiga proggies put Linux on your machine and use it for developing with a x-compiler. > Oh well, back to my PC for a bit of Quake 3 mod making in Visual Dev > Studio... MMMMmmmmMMmmmmm... Yep, and I'll go back playin' Soul Calibur on my Dreamcast which costs a fraction of a PeeCee cost and has much better gfx hardware :-) Gabriele -- A computer is as good as the one who uses it -- ------------------------------------------------------------------------ GET A NEXTCARD VISA, in 30 seconds! Get rates as low as 0.0% Intro APR and no hidden fees. Apply NOW! http://click.egroups.com/1/975/1/_/451227/_/953390382/ ------------------------------------------------------------------------ @ENDNODE @NODE "Mail_242" " Re: Single precision math functions " From: Gabriele Svelto Date: 03-18-00 15:43:43 Subject: @{b}Re: Single precision math functions@{ub} From: Gabriele Svelto > From: Gabriele Svelto > > Hello 'verybody, > does anybody know how do I do single precision math functions under > VBCC in native mode (i.e. whithout using mathieee* libraries) both for > 68k and vbcc processors? I know that standard ANSI functions are Sorry guys, I meant 68k and *ppc* processors > double precision so I cannot use them (too slow...). Thanx in advance > > Gabriele > ------------------------------------------------------------------------ DON'T HATE YOUR RATE! Get a NextCard Visa, in 30 seconds! Get rates as low as 0.0% Intro or 9.9% Fixed APR and no hidden fees. Apply NOW! http://click.egroups.com/1/2120/1/_/451227/_/953390673/ ------------------------------------------------------------------------ @ENDNODE @NODE "Mail_243" " Re: Opening libraries... " From: Andrew Bell Date: 03-18-00 15:56:42 Subject: @{b}Re: Opening libraries...@{ub} From: Andrew Bell Hi andrewmarkwell@ukonline.co.uk, On 18-Mar-00 you wrote: RE: [amiga-c] Opening libraries... > What is UTILITYNAME? Does it work if you replace it > with "utility.library"? UTILITYNAME is the define for "utility.library", as found in utility/utility.h > >> Help! All this just to get old code to compile so that I can >> fina lly >> scrap HiSOFT! > > HiSoft rocks! The IDE is great, the code and bugs suck. :( Regards, Andrew Bell. -- +----------------------------------------+ email: mailto:andrew.ab2000@bigfoot.com web: http://www.andrewb.exl.co.uk +----------------------------------------+ Sexy men are nature's way of saying keep it up! ------------------------------------------------------------------------ GET A NEXTCARD VISA, in 30 seconds! Get rates as low as 0.0% Intro or 9.9% Fixed APR and no hidden fees. Apply NOW! http://click.egroups.com/1/937/1/_/451227/_/953417796/ ------------------------------------------------------------------------ @ENDNODE @NODE "Mail_244" " Re: Opening libraries... " From: Steven Dobbs Date: 03-18-00 16:52:02 Subject: @{b}Re: Opening libraries...@{ub} From: Steven Dobbs Hello Alexander On 18-Mar-00, you wrote: > From: Alexander Niven-Jenkins > > Hiya, > > What is wrong with this? I have used this sequence to open a library in > everything even dating back to NorthC but with StormC it chucks... > > struct Library *UtilityBase = NULL; > > if( UtilityBase = OpenLibrary( UTILITYNAME, 39 ) ) > { > // code here > if( UtilityBase ) > { > CloseLibrary( UtilityBase ); > } > } > If your problem is what I think it is, try this: (view with text styles OFF) if (UtilityBase = (struct Library *)OpenLibrary(UTILITYNAME,39)){.... ^^^^^^^^^^^^^^^^^ of course, make for certain that #define UTILITYNAME is correct. though it should be if it worked on old stuff :O) Regards ------------------------------------------------------------------------ GET A NEXTCARD VISA, in 30 seconds! Get rates as low as 0.0% Intro or 9.9% Fixed APR and no hidden fees. Apply NOW! http://click.egroups.com/1/937/1/_/451227/_/953394810/ ------------------------------------------------------------------------ @ENDNODE @NODE "Mail_245" " Using MUI with VBCC " From: "G. Burdett" Date: 03-18-00 19:04:56 Subject: @{b}Using MUI with VBCC@{ub} From: "G. Burdett" Hi, Does anyone know how to make MUI work with VBCC? I've tried getting rid of all the nasty pragmas and other compiler specific stuff in the file demo.h from the dev archive, just to see whether I can get some of the example code to compile, but I'm not having much luck with it - vbcc just hangs, even with -c on the command line. Can VBCC handle the MUI macro language alright? Is it OK to use VBCC or does MUI use C++? Cheers, Giles ------------------------------------------------------------------------ GET A NEXTCARD VISA, in 30 seconds! Get rates as low as 0.0% Intro APR and no hidden fees. Apply NOW! http://click.egroups.com/1/975/1/_/451227/_/953406430/ ------------------------------------------------------------------------ @ENDNODE @NODE "Mail_246" " watches " From: Jeffrey Barnes Date: 03-18-00 19:29:37 Subject: @{b}watches@{ub} From: Jeffrey Barnes Does any male have a casio or timex watch?? ____________________________________________________________________ Get your own FREE, personal Netscape WebMail account today at http://webmail.netscape.com. ------------------------------------------------------------------------ PERFORM CPR ON YOUR APR! Get a NextCard Visa, in 30 seconds! Get rates as low as 0.0% Intro or 9.9% Fixed APR and no hidden fees. Apply NOW! http://click.egroups.com/1/2121/1/_/451227/_/953404179/ ------------------------------------------------------------------------ @ENDNODE @NODE "Mail_247" " Re: watches " From: Alexander Niven-Jenkins Date: 03-18-00 19:34:43 Subject: @{b}Re: watches@{ub} From: Alexander Niven-Jenkins Hiya Jeffrey, On 18-Mar-00, you wrote: > Does any male have a casio or timex watch?? Huh? Yeah a Casio chronograph illumintor, best £100 I ever spent, apart from the $600 for the 2 Amiga watches :) Loyalty or what! Kind regards... Alex -- Cremlin Software - http://www.cremlinsoftware.org Founder, lead programmer - ghandi@cremlinsoftware.org Production, programmer - anj@cremlinsoftware.org ------------------------------------------------------------------------ GET A NEXTCARD VISA, in 30 seconds! Get rates as low as 0.0% Intro APR and no hidden fees. Apply NOW! http://click.egroups.com/1/975/1/_/451227/_/953405235/ ------------------------------------------------------------------------ @ENDNODE @NODE "Mail_248" " Re: Opening libraries... " From: Alexander Niven-Jenkins Date: 03-18-00 19:46:39 Subject: @{b}Re: Opening libraries...@{ub} From: Alexander Niven-Jenkins Hiya Steven, On 18-Mar-00, you wrote: >> struct Library *UtilityBase = NULL; >> if( UtilityBase = OpenLibrary( UTILITYNAME, 39 ) ) >> { >> // code here >> if( UtilityBase ) >> { >> CloseLibrary( UtilityBase ); >> } >> } Unfortunately not as OpenLibrary does return struct Library* so casting it makes no difference (I have tried). The errors storm gives are: if( UtilityBase = OpenLibrary( UTILITYNAME, 39 ) ) ^ Syntax error in expression or declarator. if( UtilityBase = OpenLibrary( UTILITYNAME, 39 ) ) ^ Assignment: left value must be L-value. if( UtilityBase ) ^ Syntax error in expression or declarator. CloseLibrary( UtilityBase ); ^ Syntax error in expression or declarator. ^ is where it places the cursor, it's as though Storm doesn't like struct Library*? I am completely confused by this, like I said I have always opened libaries this way. > of course, make for certain that #define UTILITYNAME is correct. though it > should be if it worked on old stuff :O) This is the official thing in the official includes, I have never had trouble with them before. Kind regards... Alex -- Cremlin Software - http://www.cremlinsoftware.org Founder, lead programmer - ghandi@cremlinsoftware.org Production, programmer - anj@cremlinsoftware.org ------------------------------------------------------------------------ GET A NEXTCARD VISA, in 30 seconds! Get rates as low as 2.9% Intro or 9.9% Fixed APR and no hidden fees. Apply NOW! http://click.egroups.com/1/936/1/_/451227/_/953405234/ ------------------------------------------------------------------------ @ENDNODE @NODE "Mail_249" " Re: Using MUI with VBCC " From: Hynek Schlawack Date: 03-18-00 20:40:27 Subject: @{b}Re: Using MUI with VBCC@{ub} From: Hynek Schlawack Am 18-Mär-00 schrieb G. Burdett: > Does anyone know how to make MUI work with VBCC? I've tried getting rid of Just use the DLibs and everything works (I'm developing a program under vbcc and MUI and it works perfectly). > all the nasty pragmas and other compiler specific stuff in the file demo.h > from the dev archive, just to see whether I can get some of the example > code to compile, but I'm not having much luck with it - vbcc just hangs, > even with -c on the command line. Hangs? Has it enough stack? Does it perhaps just take a little bit long :-)? Have you compiled other programs w/o problems? > Can VBCC handle the MUI macro language alright? Yes. > Is it OK to use VBCC or does MUI use C++? Yes it is okay to use vbcc and no MUI doesn't use C++. The problem is that when the demo has been written no vbcc has existed (oficially :-)) so no support is included and it is also possible that the sources aren't 100% ANSI compliant...perhaps there are any other MUI sources? Try to use the sources from the DICE archive because I could port my DICE-MUI sources to vbcc within seconds and the problems weren't in the MUI parts. HTH, -- Hynek Schlawack ++ 7Wire Technologies ++ http://me.in-berlin.de/~hys ------------------------------------------------------------------------ DON'T HATE YOUR RATE! Get a NextCard Visa, in 30 seconds! Get rates as low as 0.0% Intro or 9.9% Fixed APR and no hidden fees. Apply NOW! http://click.egroups.com/1/2120/1/_/451227/_/953412480/ ------------------------------------------------------------------------ @ENDNODE @NODE "Mail_250" " stormamiga and __stdiowin... " From: Alexander Niven-Jenkins Date: 03-18-00 20:45:10 Subject: @{b}stormamiga and __stdiowin...@{ub} From: Alexander Niven-Jenkins Hiya, I have the following program. realmain() { cout << "hello world" << endl; } wbmain( struct WBStarup *wbs ) { realmain(); } main() { realmain(); } Of course there is more to this, however according to the stormamiga guide it automatically assigns stdin and stdout to a CLI (CON://///AUTO/CLOSE/WAIT) window. This is the functionality I used to use in HiSOFT C++ and so of course would like to use it again (so no code mods), however if run from WB 'hello world' does not get shoved to a CLI window? What I am doing wrong? Sorry to keep bothering you guys, but it is all these little niggly bits I need to sort out, the main stuff is working brilliantly, in fact with stromaimga.lib (regged) the final executable is actually smaller than the HiSOFT C++ one. All help really appreciated!!! Kind regards... Alex -- Cremlin Software - http://www.cremlinsoftware.org Founder, lead programmer - ghandi@cremlinsoftware.org Production, programmer - anj@cremlinsoftware.org ------------------------------------------------------------------------ GET A NEXTCARD VISA, in 30 seconds! Get rates as low as 0.0% Intro or 9.9% Fixed APR and no hidden fees. Apply NOW! http://click.egroups.com/1/937/1/_/451227/_/953408725/ ------------------------------------------------------------------------ @ENDNODE @NODE "Mail_251" " Re: Opening libraries... " From: Martin McKenzie Date: 03-18-00 21:20:05 Subject: @{b}Re: Opening libraries...@{ub} From: Martin McKenzie Hello Alexander, On 18-Mar-00, Alexander Niven-Jenkins wrote: > if( UtilityBase = OpenLibrary( UTILITYNAME, 39 ) ) You could probably #ifdef out the code to open/close the utility library, since Storm will automatically open/close standard libraries (if you use one of their functions) for you. -- Cheerio, Martin McKenzie Imagine support: http://www.cadtech.demon.co.uk/ ------------------------------------------------------------------------ GET A NEXTCARD VISA, in 30 seconds! Get rates as low as 2.9% Intro or 9.9% Fixed APR and no hidden fees. Apply NOW! http://click.egroups.com/1/936/1/_/451227/_/953411115/ ------------------------------------------------------------------------ @ENDNODE @NODE "Mail_252" " Re: Opening libraries... " From: David Scheibler Date: 03-18-00 21:51:31 Subject: @{b}Re: Opening libraries...@{ub} From: David Scheibler Am 18-Mär-00 schrieb Alexander Niven-Jenkins: struct UtilityBase *UtilityBase; if (UtilityBase = (struct UtilityBase *) OpenLibrary("utility.library",39)) { printf("ok\n"); } -- // MfG, David Scheibler // scheibler@germanynet.de (private) \X/ www.maxim-computer.de \X/ DScheibler@maxim-computer.de (work) ------------------------------------------------------------------------ DON'T HATE YOUR RATE! Get a NextCard Visa, in 30 seconds! Get rates as low as 0.0% Intro or 9.9% Fixed APR and no hidden fees. Apply NOW! http://click.egroups.com/1/2120/1/_/451227/_/953413116/ ------------------------------------------------------------------------ @ENDNODE @NODE "Mail_253" " Port control? " From: David Erman Date: 03-18-00 22:03:14 Subject: @{b}Port control?@{ub} From: David Erman Hiya guys! I was thinking. I know it's possible to gain quite a lot of control over the parallel port and serial port via misc.resource, but is there a) A way to gain this kind of control in a more legal manner, i.e. so that it'll work with third-party expansions. I.e. GVP IOExtender and MFCIII? b) Any way at all to gain access to the STROBE-line on the par. port? dazed & confused / Dave -- Studying Computer Sciences & Signal Processing UIN: 5361039 A4000/Apollo 040@40MHz/32+16Mb/(1.6+1+.5+.5+.3)Gb/32X Philips/M1438 RedHat Linux 6.0 -- AmigaOS 3.1 -- MacOS 8.1 ------------------------------------------------------------------------ MAXIMIZE YOUR CARD, MINIMIZE YOUR RATE! Get a NextCard Visa, in 30 seconds! Get rates as low as 0.0% Intro or 9.9% Fixed APR and no hidden fees. Apply NOW! http://click.egroups.com/1/2122/1/_/451227/_/953413398/ ------------------------------------------------------------------------ @ENDNODE @NODE "Mail_254" " Re: Opening libraries... " From: Andrew Bell Date: 03-18-00 23:44:11 Subject: @{b}Re: Opening libraries...@{ub} From: Andrew Bell Hi Alexander, On 18-Mar-00 you wrote: [amiga-c] Re: Opening libraries... > Unfortunately not as OpenLibrary does return struct Library* so > casting it makes no difference (I have tried). > > The errors storm gives are: > > if( UtilityBase = OpenLibrary( UTILITYNAME, 39 ) ) > ^ > Syntax error in expression or declarator. > > if( UtilityBase = OpenLibrary( UTILITYNAME, 39 ) ) > ^ > Assignment: left value must be L-value. > > if( UtilityBase ) > ^ > Syntax error in expression or declarator. > > CloseLibrary( UtilityBase ); > ^ > Syntax error in expression or declarator. > > ^ is where it places the cursor, it's as though Storm doesn't > like struct Library*? I am completely confused by this, like I said > I have always opened libaries this way. Just out of curiousity, is UtilityBase a local or global variable? Regards, Andrew Bell. -- +----------------------------------------+ email: mailto:andrew.ab2000@bigfoot.com web: http://www.andrewb.exl.co.uk +----------------------------------------+ Tomorrow is another day, but it'll suck too. ------------------------------------------------------------------------ GET A NEXTCARD VISA, in 30 seconds! Get rates as low as 2.9% Intro or 9.9% Fixed APR and no hidden fees. Apply NOW! http://click.egroups.com/1/936/1/_/451227/_/953430184/ ------------------------------------------------------------------------ @ENDNODE @NODE "Mail_255" " Re: Using MUI with VBCC " From: Andrew Bell Date: 03-19-00 00:10:59 Subject: @{b}Re: Using MUI with VBCC@{ub} From: Andrew Bell Hi G., On 18-Mar-00 you wrote: [amiga-c] Using MUI with VBCC > Does anyone know how to make MUI work with VBCC? I've tried getting > rid of all the nasty pragmas and other compiler specific stuff in > the file demo.h from the dev archive, just to see whether I can get > some of the example code to compile, but I'm not having much luck > with it - vbcc just hangs, even with -c on the command line. Can > VBCC handle the MUI macro language alright? > > Is it OK to use VBCC or does MUI use C++? I've compiled some large MUI based projects (one of which is 10000 lines+) on many C compilers for testing purposes, including VBCC. VBCC handled all of the MUI code without any problems. The MUI examples only cater for SAS C, GNU C and DICE compilers. To get these examples to work under VBCC, you *will* need to make a lot of source modifications. Don't try to dive into the deep end first, unless you know what you're doing. Instead, write some small test programs and compile these first. This will allow you to fix each problem one at a time. Remember VBCC is a very strict compiler. Regards, Andrew Bell. -- +----------------------------------------+ email: mailto:andrew.ab2000@bigfoot.com web: http://www.andrewb.exl.co.uk +----------------------------------------+ If picketing didn't exist some kids would never exercise. ------------------------------------------------------------------------ DON'T HATE YOUR RATE! Get a NextCard Visa, in 30 seconds! Get rates as low as 0.0% Intro or 9.9% Fixed APR and no hidden fees. Apply NOW! http://click.egroups.com/1/2120/1/_/451227/_/953430184/ ------------------------------------------------------------------------ @ENDNODE @NODE "Mail_256" " "Segmentation Fault" sort of OT " From: Steven Dobbs Date: 03-19-00 02:53:39 Subject: @{b}"Segmentation Fault" sort of OT@{ub} From: Steven Dobbs Hello People, I've been trying to use a Linux PC as a developer machine for my game. I was wondering if somebody could explain to me what a segmentation failt is during run time, I'm using corel linux if that means anything. any help would be very much appreciated :O) Regards Steven Dobbs ------------------------------------------------------------------------ MAXIMIZE YOUR CARD, MINIMIZE YOUR RATE! Get a NextCard Visa, in 30 seconds! Get rates as low as 0.0% Intro or 9.9% Fixed APR and no hidden fees. Apply NOW! http://click.egroups.com/1/2122/1/_/451227/_/953430907/ ------------------------------------------------------------------------ @ENDNODE @NODE "Mail_257" " Re: Re: Single precision math functions " From: Volker Barthelmann Date: 03-19-00 04:09:21 Subject: @{b}Re: Re: Single precision math functions@{ub} From: Volker Barthelmann On Sat, 18 Mar 2000, Gabriele Svelto wrote: > From: Gabriele Svelto > > > Hello 'verybody, > > does anybody know how do I do single precision math functions under > > VBCC in native mode (i.e. whithout using mathieee* libraries) both for > > 68k and vbcc processors? I know that standard ANSI functions are > > Sorry guys, I meant 68k and *ppc* processors > > > double precision so I cannot use them (too slow...). Thanx in advance I'm not sure I understand your question. If you want to use functions like sin() with single-precision using the FPU then you can look at the math.h include files which have inline-assembly versions of some of the standard math-functions. Note, however, that the PPC as well as 68040/060 do not have hardware-support for most of these functions. vbcc comes with the Motorola routines which are very fast, even though they are double-precision. You probably should look for another solution (rough approximations/table lookups/different ways of calculating?) if these are really too slow for your needs. Sorry, if I have misunderstood your question. Volker ------------------------------------------------------------------------ DON'T HATE YOUR RATE! Get a NextCard Visa, in 30 seconds! Get rates as low as 0.0% Intro or 9.9% Fixed APR and no hidden fees. Apply NOW! http://click.egroups.com/1/2120/1/_/451227/_/953436407/ ------------------------------------------------------------------------ @ENDNODE @NODE "Mail_258" " Re: Re: Inline " From: Duane McDonnell Date: 03-19-00 06:36:39 Subject: @{b}Re: Re: Inline@{ub} From: Duane McDonnell >From: Ilkka Lehtoranta >On 17-Maa-00, Bart King wrote: >BK> Something that's been bugging me for ages is whether or not it is >BK> possible to use inline asm code in C code on Amiga compilers? It seems >BK> to be a waste of time optimising small parts of C code in asm due to the >BK> overheads gained from having three lines of asm in a separate object. >I don't know why you want inline asm code. Very likely users can't see any >difference in speed. Not even if they are using a vanilla A1200. >Optimizing C code in asm _is_ a waste of time. Rubbish, plenty of applications and games are C/asm hybrids, probably including the very link libraries you use in your own C development and possibly even the C compiler itself (DICE for example). You can often achieve large increases in speed if you optimize inner loops for such things as rendering code. Besides speed, there are cases where C simply isn't capable of doing what you want and asm must be used instead. To the original poster: if you're using SAS/C the emit support might get you out of trouble for simple/small bits of code. I believe the old Aztec compiler supported #asm too (not sure if it branced or inlined though). I haven't used it on the Amiga, but gcc might offer the features you need but you'll have to learn the funny format :-) ------------------------------------------------------------------------ MAXIMIZE YOUR CARD, MINIMIZE YOUR RATE! Get a NextCard Visa, in 30 seconds! Get rates as low as 0.0% Intro or 9.9% Fixed APR and no hidden fees. Apply NOW! http://click.egroups.com/1/2122/1/_/451227/_/953381492/ ------------------------------------------------------------------------ @ENDNODE @NODE "Mail_259" " Re: "Segmentation Fault" sort of OT " From: Stephen Illingworth Date: 03-19-00 09:52:10 Subject: @{b}Re: "Segmentation Fault" sort of OT@{ub} From: Stephen Illingworth Hello Steven On 19-Mar-00, you wrote: > From: Steven Dobbs > > Hello People, > > I've been trying to use a Linux PC as a developer machine for my game. I > was wondering if somebody could explain to me what a segmentation failt is > during run time, I'm using corel linux if that means anything. any help > would be very much appreciated :O) > A segmentation fault occurs when a process attempts to access a invalid virtual memory address or attempts to access memory that it does not have permission to use. This is a generic feature of all UNIXes BTW. Look in your code to see if you're overrunning an array or using memory that has already been freed. Steve ------------------------------------------------------------------------ DON'T HATE YOUR RATE! Get a NextCard Visa, in 30 seconds! Get rates as low as 0.0% Intro or 9.9% Fixed APR and no hidden fees. Apply NOW! http://click.egroups.com/1/2120/1/_/451227/_/953459682/ ------------------------------------------------------------------------ @ENDNODE @NODE "Mail_260" " Re: "Segmentation Fault" sort of OT " From: David Erman Date: 03-19-00 10:06:48 Subject: @{b}Re: "Segmentation Fault" sort of OT@{ub} From: David Erman Hi Steven, In a mail of 19-Mar-00, you wrote: SD> I've been trying to use a Linux PC as a developer machine for my game. I was SD> wondering if somebody could explain to me what a segmentation failt is SD> during run time, I'm using corel linux if that means anything. any help SD> would be very much appreciated :O) IIRC from my CS class, it means you're mucking about in memory that isn't yours. Could be caused by a multitude of coding errors, f.ex. not checking for a null pointer, doing a for(i=0;i<=10;i++) instead of a for(i=0;i<10;i++) etc. Yours, Dave -- Studying Computer Sciences & Signal Processing UIN: 5361039 A4000/Apollo 040@40MHz/32Mb/1.6Gb/4X Mitsumi/M1438 RedHat Linux 6.0 -- AmigaOS 3.1 -- MacOS 8.1 ------------------------------------------------------------------------ GET A NEXTCARD VISA, in 30 seconds! Get rates as low as 0.0% Intro or 9.9% Fixed APR and no hidden fees. Apply NOW! http://click.egroups.com/1/937/1/_/451227/_/953456634/ ------------------------------------------------------------------------ @ENDNODE @NODE "Mail_261" " Re: Opening libraries... " From: Alexander Niven-Jenkins Date: 03-19-00 10:16:04 Subject: @{b}Re: Opening libraries...@{ub} From: Alexander Niven-Jenkins Hiya Andrew, On 18-Mar-00, you wrote: > Just out of curiousity, is UtilityBase a local or global variable? Global. But see othe msg, a small project which literally opens the library on it's own works, so it must be parts of my other code that's causing it to go all funny. Kind regards... Alex -- Cremlin Software - http://www.cremlinsoftware.org Founder, lead programmer - ghandi@cremlinsoftware.org Production, programmer - anj@cremlinsoftware.org ------------------------------------------------------------------------ GET A NEXTCARD VISA, in 30 seconds! Get rates as low as 0.0% Intro or 9.9% Fixed APR and no hidden fees. Apply NOW! http://click.egroups.com/1/937/1/_/451227/_/953457498/ ------------------------------------------------------------------------ @ENDNODE @NODE "Mail_262" " Re: Editors with Syntax highlighting " From: Olivier Fabre Date: 03-19-00 13:03:02 Subject: @{b}Re: Editors with Syntax highlighting@{ub} From: Olivier Fabre On 15-mars-00, Dominic Clifton, you wrote: > I was reading a message about editors and thought to myself "I wonder what > editors there are now on the Amiga that support Syntax highlighting" [...] > What other editors are there out there ? I still use TurboText, it's great, > small, and very quick (except for loading large files). CygnusEd 4 is the best I found. Unfortunately its menues are not very logically organised & quite big & not configurable, the keyboard shortcuts are also not configurable and not adapted to an french keyboard, and most importantly it doesn't support syntax highlighting and automatic indentation. I really wish to see (and buy) an improved version. On the other hand, I recently installed StormC 3 (Dev CD), and the Storm Editor is not bad at all, with colours and indentation... Unfortunately, it looks like some settings have to be set every time you create a new file... Annoying. -- // Olivier Fabre - StAn_ on IRCnet - http://off.free.fr \X/ A1200 Apollo 1260 Pixel64 APNet 4.3GB E/Box OS3.5 ------------------------------------------------------------------------ GET A NEXTCARD VISA, in 30 seconds! Get rates as low as 0.0% Intro or 9.9% Fixed APR and no hidden fees. Apply NOW! http://click.egroups.com/1/937/1/_/451227/_/953502643/ ------------------------------------------------------------------------ @ENDNODE @NODE "Mail_263" " Re: stormamiga and __stdiowin... " From: Allan Odgaard Date: 03-19-00 23:43:33 Subject: @{b}Re: stormamiga and __stdiowin...@{ub} From: Allan Odgaard On 18-Mar-00, Alexander Niven-Jenkins wrote: > [...] according to the stormamiga guide it automatically assigns stdin > and stdout to a CLI (CON://///AUTO/CLOSE/WAIT) window. Only if you don't supply your own wbmain()-function. Regards Allan -- URL: http://www.diku.dk/students/duff/ People will pay *big bucks* for the luxury of ignorance. ------------------------------------------------------------------------ PERFORM CPR ON YOUR APR! Get a NextCard Visa, in 30 seconds! Get rates as low as 0.0% Intro or 9.9% Fixed APR and no hidden fees. Apply NOW! http://click.egroups.com/1/2121/1/_/451227/_/953506370/ ------------------------------------------------------------------------ @ENDNODE @NODE "Mail_264" " Re: Opening libraries... " From: Allan Odgaard Date: 03-19-00 23:50:38 Subject: @{b}Re: Opening libraries...@{ub} From: Allan Odgaard On 18-Mar-00, Alexander Niven-Jenkins wrote: > What is wrong with this? I have used this sequence to open a library in > everything even dating back to NorthC but with StormC it chucks... > struct Library *UtilityBase = NULL; > if( UtilityBase = OpenLibrary( UTILITYNAME, 39 ) ) > [...] a) I don't know why you want to keep the code when StormC will open the lib for you. b) It works fine here. c) You could check if one of the spaces should accidentally be a nbsp, this is known to cause weird compiler error messages. Regards Allan -- URL: http://www.diku.dk/students/duff/ I am Windows of Borg. Prepare... oops, out of memory! ------------------------------------------------------------------------ GET A NEXTCARD VISA, in 30 seconds! Get rates as low as 0.0% Intro APR and no hidden fees. Apply NOW! http://click.egroups.com/1/975/1/_/451227/_/953506371/ ------------------------------------------------------------------------ @ENDNODE @NODE "Mail_265" " Opening Libraries " From: "Alastair M. Robinson" Date: 03-20-00 01:04:47 Subject: @{b}Opening Libraries@{ub} From: "Alastair M. Robinson" Hi, > Global. But see othe msg, a small project which literally opens the > library on it's own works, so it must be parts of my other code that's > causing it to go all funny. Assuming you don't actually need to dereference it, have you tried declaring UtilityBase as a void * instead of a struct Library * ? Another thing to check is your includes - or specifically, the order of them: Strict ANSI compilers can throw up strange errors (with VBCC usually just warnings) if a structure is referenced in a function definition before the structure itself is declared. Make sure you're including (for that matter, all system includes) before you start with or . Another possibility: Make sure all the spaces in your source are normal spaces, and not non-breaking spaces (alt-space). I sometimes type alt-space accidentally (on this PC-keyboard), and have to remove them before the file will compile. Hope these three suggestions help, -- Alastair M. Robinson, email:blackfive@fakenhamweb.co.uk Amiga 4000/030, 18Meg RAM, 4.3Gig HD, Mustek ScanExpress 6000SP C, Assembler (68k, PIC16C84, ATMEL AVR), PostScript, ARexx, AMOSPro(!), AmigaDOS Techno is to music what Etch-a-Sketch is to art! ------------------------------------------------------------------------ GET A NEXTCARD VISA, in 30 seconds! Get rates as low as 0.0% Intro or 9.9% Fixed APR and no hidden fees. Apply NOW! http://click.egroups.com/1/937/1/_/451227/_/953594101/ ------------------------------------------------------------------------ @ENDNODE @NODE "Mail_266" " RE:Opening Libraries " From: roderick_maclean@agilent.com Date: 03-20-00 11:24:28 Subject: @{b}RE:Opening Libraries@{ub} From: roderick_maclean@agilent.com >> struct UtilityBase *UtilityBase; > >> if (UtilityBase = (struct UtilityBase *) >> OpenLibrary("utility.library",39)) { >> printf("ok\n"); >> } > > Thanks, however the below works. So what's the difference > between struct > UtilityBase and struct Library??? Well... what is the difference between ExecBase, GfxBase and all the others and struct Library? My guess (and it is only a guess) is that at some point someone was planning to make 'struct Library' the superclass of all the other Library structures. Anyway, I've just had to get into the habit of saying struct somelibBase *somelibBase; if(somelibBase = (struct somelibBase *)OpenLibrary("somelib.library", SOMELIB_VERSION)) { /* do something */ { > > // testcode.cpp > > #include > #include > #include > > struct Library *UtilityBase = NULL; > Are you sure that you're not redefining UtilityBase here? IIRC, my utility/utility.h defines UtilityBase for me... Rod -- Rod MacLean | Roderick_MacLean@agilent.com Developer - Agilent Technologies | +44 (0131) 331-7304 ** My views do not necessarily reflect the views of Agilent! ** ------------------------------------------------------------------------ PERFORM CPR ON YOUR APR! Get a NextCard Visa, in 30 seconds! Get rates as low as 0.0% Intro or 9.9% Fixed APR and no hidden fees. Apply NOW! http://click.egroups.com/1/2121/1/_/451227/_/953547895/ ------------------------------------------------------------------------ @ENDNODE @NODE "Mail_267" " Re: Inline " From: Ilkka Lehtoranta Date: 03-20-00 12:13:55 Subject: @{b}Re: Inline@{ub} From: Ilkka Lehtoranta Hello Duane On 19-Maa-00, you wrote: DM>> I don't know why you want inline asm code. Very likely users can't see DM>> any difference in speed. Not even if they are using a vanilla A1200. DM> DM>> Optimizing C code in asm _is_ a waste of time. DM> DM> Rubbish, plenty of applications and games are C/asm hybrids, probably DM> including the very link libraries you use in your own C development DM> and possibly even the C compiler itself (DICE for example). You can DM> often achieve large increases in speed if you optimize inner loops DM> for such things as rendering code. Besides speed, there are cases DM> where C simply isn't capable of doing what you want and asm must DM> be used instead. Well, ok, asm optimizations may make sense sometimes, but for example SAS/C produces very good asm code. I have seen SAS/C generating far better code than many asm coders do. Some optimizations are even too complicated for any asm coder but piece of cake for SAS/C. On my 040/25Mhz, if C routine doesn't go fast enough then probably hand written asm code can't help enough. (Assuming routine is well designed.) However, I'm still creating my apps in asm rather than in C since I can write 1000+ lines of asm code when I'm still writing my first C line ;-) It is so easy... But I have not seen real difference with my hand written asm to C equivalents today. ------------------------------------------------------------------------ PERFORM CPR ON YOUR APR! Get a NextCard Visa, in 30 seconds! Get rates as low as 0.0% Intro or 9.9% Fixed APR and no hidden fees. Apply NOW! http://click.egroups.com/1/2121/1/_/451227/_/953551352/ ------------------------------------------------------------------------ @ENDNODE @NODE "Mail_268" " RE: Inline " From: Jesper Svennevid Date: 03-20-00 12:49:03 Subject: @{b}RE: Inline@{ub} From: Jesper Svennevid > From: Bart King > > Hi. > > Something that's been bugging me for ages is whether or not it is > possible to use inline asm code in C code on Amiga compilers? > It seems to be a waste of time optimising small parts of C code in asm > due to the overheads gained from having three lines of asm in a separate object. > > Like MSVC: > > for(int blah = 0; blah < 20; blah++) > { > _asm > { > ... > } > } > > I know EGCS supports inline asm from Linux, but unsure about StormC > (manual is really useless), SAS/C, etc... The only amiga C-compilers I know of that supports inlining assembler-code is GCC(like you said), and VBCC. (And IMHO, to all the "we-wont-use-asm" people out there, optimizing routines in assembler is a must sometimes even today, as the C-compilers on amiga have a tendency on using memory way too much.) -- // Jesper Svennevid : jesper@fxrealm.com Software Developer at FXRealm Studios http://www.fxrealm.com/ ------------------------------------------------------------------------ DON'T HATE YOUR RATE! Get a NextCard Visa, in 30 seconds! Get rates as low as 0.0% Intro or 9.9% Fixed APR and no hidden fees. Apply NOW! http://click.egroups.com/1/2120/1/_/451227/_/953699719/ ------------------------------------------------------------------------ @ENDNODE @NODE "Mail_269" " RE: Re: What PPC compiler (was: What compiler....) " From: Jesper Svennevid Date: 03-20-00 12:53:29 Subject: @{b}RE: Re: What PPC compiler (was: What compiler....)@{ub} From: Jesper Svennevid > From: Gavin Kinsey > > gcc is much easier to set up now, all you have to do is > download the archive from my site > and unlzx to GG:. > If > you have gcc working for 68k then that should be enough to > get it working for ppc, it might work on its own too but > I'm not sure about that. I checked those compiler-cores last week, and it crashed while compiling the first source I tried, so I switched back to 2.91.66 :) Is there perhaps some incompatibility with my includes (I use ones that are align-fixed), or stack problems(I run everything I use with 250kB stack)? -- // Jesper Svennevid : jesper@fxrealm.com Software Developer at FXRealm Studios http://www.fxrealm.com/ ------------------------------------------------------------------------ PERFORM CPR ON YOUR APR! Get a NextCard Visa, in 30 seconds! Get rates as low as 0.0% Intro or 9.9% Fixed APR and no hidden fees. Apply NOW! http://click.egroups.com/1/2121/1/_/451227/_/953699728/ ------------------------------------------------------------------------ @ENDNODE @NODE "Mail_270" " Re: What PPC compiler (was: What compiler....) " From: Fritsch Alexander Date: 03-20-00 13:25:05 Subject: @{b}Re: What PPC compiler (was: What compiler....)@{ub} From: Fritsch Alexander > -----Ursprüngliche Nachricht----- > Von: Gavin Kinsey [SMTP:gavin@kinsey.u-net.com] > Gesendet am: Freitag, 17. März 2000 08:06 > An: amiga-c@onelist.com > Betreff: [amiga-c] Re: What PPC compiler (was: What compiler....) > > From: Gavin Kinsey > ...... > gcc is much easier to set up now, all you have to do is download the > archive > from my site and unlzx to GG:. If > you have gcc working for 68k then that should be enough to get it working > for ppc, it might work on its own too but I'm not sure about that. ...... > Gavin Kinsey > I have the GeekGadgets CD Snapshot May 98 installed. I will try your archive today. Is there really all I need to compile for PPC? Can I use Amiga-specific functions or stdio-functions? I compile my project currently with SAS/C for 68k and for PPC. The PPC-program works fine with the ppclibemu under WarpOS. Could you please give me an example for correct compiler invocation? For my project (an emulator) I need the best speed possible, regardless of compilingtime or executable size. If my project consist of emul.cpp, bios.cpp and hardware.cpp, what would I have to type? Kind regards Alexander Fritsch ------------------------------------------------------------------------ GET A NEXTCARD VISA, in 30 seconds! Get rates as low as 0.0% Intro APR and no hidden fees. Apply NOW! http://click.egroups.com/1/975/1/_/451227/_/953555059/ ------------------------------------------------------------------------ @ENDNODE @NODE "Mail_271" " Re: Re: Inline " From: Paul Hill Date: 03-20-00 15:08:35 Subject: @{b}Re: Re: Inline@{ub} From: Paul Hill Ilkka Lehtoranta wrote: > Hello Duane > On 19-Maa-00, you wrote: > > DM>> I don't know why you want inline asm code. Very likely users can't see > DM>> any difference in speed. Not even if they are using a vanilla A1200. > DM> > DM>> Optimizing C code in asm _is_ a waste of time. > DM> > DM> Rubbish, plenty of applications and games are C/asm hybrids, probably > DM> including the very link libraries you use in your own C development > DM> and possibly even the C compiler itself (DICE for example). You can > DM> often achieve large increases in speed if you optimize inner loops > DM> for such things as rendering code. Besides speed, there are cases > DM> where C simply isn't capable of doing what you want and asm must > DM> be used instead. > > Well, ok, asm optimizations may make sense sometimes, but for example > SAS/C produces very good asm code. I have seen SAS/C generating > far better code than many asm coders do. Some optimizations are even > too complicated for any asm coder but piece of cake for SAS/C. > > On my 040/25Mhz, if C routine doesn't go fast enough then probably > hand written asm code can't help enough. (Assuming routine is well > designed.) There are other reasons for using inline assembler. One example is when you need a so-called 'atomic' instruction. i.e. where you need to perform an operation in 1 CPU instruction. One example of this is a semaphore locking system. You must set the semaphore flag in 1 instruction or there is the chance another task (or interrupt) could change the flag at the same time. Also, on the Amiga there's another reason - hardware registers. Some of these cannot be read from so you need to be sure what the code is doing (e.g. move.l #0 instead of clr.l). Having said all that, I personally stick my assembler code in separate source files and link together with the C code. It's more portable too (and I usually have a C equivalent of each asm function). -- Paul ------------------------------------------------------------------------ GET A NEXTCARD VISA, in 30 seconds! Get rates as low as 0.0% Intro or 9.9% Fixed APR and no hidden fees. Apply NOW! http://click.egroups.com/1/937/1/_/451227/_/953631320/ ------------------------------------------------------------------------ @ENDNODE @NODE "Mail_272" " ghostscript docs.. " From: Kevin Bewley Date: 03-20-00 17:51:28 Subject: @{b}ghostscript docs..@{ub} From: "Kevin Bewley" Hi, I've recently installed ghostscript (5.10) on my miggy from Aminet. But I don't seem to have the docs. Does anyone know which archive they're in so I can try downloading it again. If they're in the 3.2Mb source one (as I fear and cannot handle whole, only in 1.4Mb chunks!) could someone just lha up the docs for me and eMail the file to me? Thanks, Kevin -------- A1200 Kick 3.1 '060/50MHz + SCSI 50Mb RAM 1.3Gb Hitachi HD 2x SCSI CD-ROM ------------------------------------------------------------------------ MAXIMIZE YOUR CARD, MINIMIZE YOUR RATE! Get a NextCard Visa, in 30 seconds! Get rates as low as 0.0% Intro or 9.9% Fixed APR and no hidden fees. Apply NOW! http://click.egroups.com/1/2122/1/_/451227/_/953571094/ ------------------------------------------------------------------------ @ENDNODE @NODE "Mail_273" " Re: Single precision math functions " From: Gabriele Svelto Date: 03-20-00 19:36:47 Subject: @{b}Re: Single precision math functions@{ub} From: Gabriele Svelto Hi Volker On 19-Mar-00, you wrote: > From: Volker Barthelmann > > I'm not sure I understand your question. If you want to use functions > like sin() with single-precision using the FPU then you can look at > the math.h include files which have inline-assembly versions of some > of the standard math-functions. > > Note, however, that the PPC as well as 68040/060 do not have > hardware-support for most of these functions. vbcc comes with the > Motorola routines which are very fast, even though they are > double-precision. I knew that on 040s & 060s the instructions where done using multiple basic functions although I didn't knew that it was the same on PPCs (well, it was obvious anyhow 'cos they're RISCs). In fact I don't know exactly how single-precison instructions can be faster than double- ones, I just based my ideas on the fact that mathieeesing* libraries where faster than mathieeedoub* ones and I supposed that it must have been the same in hardware. Since I'm working on a 3d modeller I needed fast functions for visualizing objects and I don't believe that an user needs more than single precision floats for modelling objects (although nobody knows...) so I went for single precision. But if double- precision functions are fast enough (and I trust your compiler's code for being faaast) I'll better jump to double. > You probably should look for another solution (rough > approximations/table lookups/different ways of calculating?) if these > are really too slow for your needs. I probably simply drop my idea on single-precision funcs, it looks like it's too difficult to implement and I have still a *lot* of work to do on other things (amongst everything the interface). BTW the code from Motorola is avaible to the public? I am really interested on approximating transcendental functions with normal operations (w/o using Taylor/McLaurrin formules). > Sorry, if I have misunderstood your question. > > Volker You haven't ;-) thanks (I was fearing to get no answer at all :-) ) Gabriele ------------------------------------------------------------------------ MAXIMIZE YOUR CARD, MINIMIZE YOUR RATE! Get a NextCard Visa, in 30 seconds! Get rates as low as 0.0% Intro or 9.9% Fixed APR and no hidden fees. Apply NOW! http://click.egroups.com/1/2122/1/_/451227/_/953577504/ ------------------------------------------------------------------------ @ENDNODE @NODE "Mail_274" " conio.h? " From: Ilkka Lehtoranta Date: 03-20-00 23:27:13 Subject: @{b}conio.h?@{ub} From: Ilkka Lehtoranta Where can I obtain conio.h? StormC nor SAS/C have no such file... Or where I can find a list of ANSI codes for gotoxy() etc. things? ------------------------------------------------------------------------ GET A NEXTCARD VISA, in 30 seconds! Get rates as low as 0.0% Intro or 9.9% Fixed APR and no hidden fees. Apply NOW! http://click.egroups.com/1/937/1/_/451227/_/953591676/ ------------------------------------------------------------------------ @ENDNODE @NODE "Mail_275" " Re: ghostscript docs.. " From: Gabriele Svelto Date: 03-21-00 00:28:37 Subject: @{b}Re: ghostscript docs..@{ub} From: Gabriele Svelto Hi Kevin On 20-Mar-00, you wrote: > From: "Kevin Bewley" > > Hi, > > I've recently installed ghostscript (5.10) on my miggy from Aminet. > But I don't seem to have the docs. Does anyone know which archive > they're in so I can try downloading it again. If they're in the 3.2Mb > source one (as I fear and cannot handle whole, only in 1.4Mb chunks!) > could someone just lha up the docs for me and eMail the file to me? > > Thanks, > Kevin > > -------- > A1200 Kick 3.1 '060/50MHz + SCSI 50Mb RAM > 1.3Gb Hitachi HD 2x SCSI CD-ROM It looks like you're downloading from a PeeCee on hi-density disks isn't it? I dunno where are ghostscript docs but I can send you a veeery useful windoze proggie which splits files into multiple chunks (you can then rejoin the files on your miggie with the command 'join'), I can send it to you if it could be useful. Ciao Gabriele ------------------------------------------------------------------------ GET A NEXTCARD VISA, in 30 seconds! Get rates as low as 0.0% Intro or 9.9% Fixed APR and no hidden fees. Apply NOW! http://click.egroups.com/1/937/1/_/451227/_/953661584/ ------------------------------------------------------------------------ @ENDNODE @NODE "Mail_276" " What is reentrant code? " From: Gabriele Svelto Date: 03-21-00 00:32:40 Subject: @{b}What is reentrant code?@{ub} From: Gabriele Svelto Hi everybody, I am just curious about that, what is exactly reentrant code? Everybody talks about it but I haven't yet found what is it ! See ya Gabriele ------------------------------------------------------------------------ GET A NEXTCARD VISA, in 30 seconds! Get rates as low as 0.0% Intro or 9.9% Fixed APR and no hidden fees. Apply NOW! http://click.egroups.com/1/937/1/_/451227/_/953661586/ ------------------------------------------------------------------------ @ENDNODE @NODE "Mail_277" " Re: What PPC compiler (was: What compiler....) " From: Gavin Kinsey Date: 03-21-00 00:39:40 Subject: @{b}Re: What PPC compiler (was: What compiler....)@{ub} From: Gavin Kinsey I received a message from Fritsch Alexander, a reply follows, >> gcc is much easier to set up now, all you have to do is download the >> archive >> from my site and unlzx to GG:. If >> you have gcc working for 68k then that should be enough to get it working >> for ppc, it might work on its own too but I'm not sure about that. >> > I have the GeekGadgets CD Snapshot May 98 installed. I will try your > archive today. > Is there really all I need to compile for PPC? Can I use Amiga-specific > functions or stdio-functions? I compile my project currently with SAS/C > for 68k and for PPC. The PPC-program works fine with the ppclibemu under > WarpOS. I am pretty sure I put everything needed in there. I haven't tried compiling stuff that uses a lot of AmigaOS calls but it should work. I have come across a problem with stdio though, it doesn't like using relative paths to open files in parent directories, for example fp=fopen("/test","r") fails. Using the full path of the file or not accessing dirs above the current works fine though. > Could you please give me an example for correct compiler invocation? > For my project (an emulator) I need the best speed possible, regardless of > compilingtime or executable size. ppc-amigaos-gcc -V 2.95.1 xxx.c elf2exe2 a.out ProgName Will produce a WarpUp exe, for optimisation I use "-s -O2 -fomit-frame-pointer -fstrength-reduce -mmultiple -mstring -mcpu=604e", the last option may need changing to 603e to run on A1200 cards :-). You can also produce for PowerUp by adding "-pup -o ProgName" and replacing the elf2exe2 with "Protect ProgName +e". > If my project consist of emul.cpp, bios.cpp and hardware.cpp, what > would I have to type? Aargh! I just checked and although there is the cc1plus compiler in the archive the C++ includes and libs are missing. Actually, the libs don't exist on my system either so I shall have to compile them first. Bugger! Wait a couple of days and I'll update my website with the new files. -- I am Circuit Designer of Borg. Resistance is V/I. You will be simulated. Gavin Kinsey ------------------------------------------------------------------------ DON'T HATE YOUR RATE! Get a NextCard Visa, in 30 seconds! Get rates as low as 0.0% Intro or 9.9% Fixed APR and no hidden fees. Apply NOW! http://click.egroups.com/1/2120/1/_/451227/_/953753711/ ------------------------------------------------------------------------ @ENDNODE @NODE "Mail_278" " Re: Re: Single precision math functions " From: Volker Barthelmann Date: 03-21-00 03:31:32 Subject: @{b}Re: Re: Single precision math functions@{ub} From: Volker Barthelmann On Mon, 20 Mar 2000, Gabriele Svelto wrote: > I knew that on 040s & 060s the instructions where done using multiple > basic functions although I didn't knew that it was the same on PPCs > (well, it was obvious anyhow 'cos they're RISCs). In fact I don't know > exactly how single-precison instructions can be faster than double- > ones, I just based my ideas on the fact that mathieeesing* libraries > where faster than mathieeedoub* ones and I supposed that it must have > been the same in hardware. I think single-precision functions could be a bit faster (on some FPUs single-precision arithmetic is faster and you might be able to use polynomials of lower degree for approximation). I'd guess the benefit wouldn't be that high, though. > I probably simply drop my idea on single-precision funcs, it looks like > it's too difficult to implement and I have still a *lot* of work to do > on other things (amongst everything the interface). Yes, I'd save such optimizations until everything else is done. Probably they won't be necessary. Also, the best way of optimizing transcendental funtions is to call them as infrequently as possible. > BTW the code from Motorola is avaible to the public? For 68040/060 the source is available (e.g. in the vbccm68ksrc.lha archive). For PPC I'm not sure. It may be available only as object code. > I am really interested on > approximating transcendental functions with normal operations (w/o using > Taylor/McLaurrin formules). Good luck with learning from the Motorola sources. :-) Volker ------------------------------------------------------------------------ GET A NEXTCARD VISA, in 30 seconds! Get rates as low as 0.0% Intro APR and no hidden fees. Apply NOW! http://click.egroups.com/1/975/1/_/451227/_/953607290/ ------------------------------------------------------------------------ @ENDNODE @NODE "Mail_279" " RE: conio.h " From: roderick_maclean@agilent.com Date: 03-21-00 11:28:56 Subject: @{b}RE: conio.h@{ub} From: roderick_maclean@agilent.com > Subject: conio.h? > > Where can I obtain conio.h? StormC nor SAS/C have no such file... > Or where I can find a list of ANSI codes for gotoxy() etc. things? conio.h doesn't exist on non-PC C compilers. gotoxy() can be emulated with escape codes. There is a full list of them on Aminet. Look at this in a text editor: http://wuarchive.wustl.edu/aminet/dirs/aminet/dev/misc/AmigaANSI.txt You can copy the escape codes you need into an include of your own. (... call it conio.h if you like :) Rod -- Rod MacLean | Roderick_MacLean@agilent.com Developer - Agilent Technologies | +44 (0131) 331-7304 ** My views do not necessarily reflect the views of Agilent! ** ------------------------------------------------------------------------ GET A NEXTCARD VISA, in 30 seconds! Get rates as low as 2.9% Intro or 9.9% Fixed APR and no hidden fees. Apply NOW! http://click.egroups.com/1/936/1/_/451227/_/953634553/ ------------------------------------------------------------------------ @ENDNODE @NODE "Mail_280" " Re: Opening libraries... " From: Bart King Date: 03-21-00 13:12:51 Subject: @{b}Re: Opening libraries...@{ub} From: Bart King Alexander Niven-Jenkins wrote: > struct Library *UtilityBase = NULL; > > if( UtilityBase = OpenLibrary( UTILITYNAME, 39 ) ) It's a problem with StormC's includes/link libraries. Just use: extern struct Library *UtilityBase; Works fine for me. You shouldn't need v39 of UtilityBase anyway - check v39 with something more important like intuition.library or exec.library. -- Bart King Developer http://www.bartman.demon.co.uk (not up yet - soon) http://www.fscked.net/fubar/ ICQ: 18178781 ------------------------------------------------------------------------ GET A NEXTCARD VISA, in 30 seconds! Get rates as low as 0.0% Intro or 9.9% Fixed APR and no hidden fees. Apply NOW! http://click.egroups.com/1/937/1/_/451227/_/953677141/ ------------------------------------------------------------------------ @ENDNODE @NODE "Mail_281" " Re: conio.h? " From: Tak Tang Date: 03-21-00 14:21:47 Subject: @{b}Re: conio.h?@{ub} From: "Tak Tang" > Where can I obtain conio.h? StormC nor SAS/C have no such file... > Or where I can find a list of ANSI codes for gotoxy() etc. things? When you install SAS/C, it should install the C++ header files into the subdirectory "cxxinclude". You will find "conio.h" in there. You might have to end the filename with ".cpp" so it knows to use the C++ extensions. I imagine they are also included with StormC too. For tricks with the console device, such as "gotoxy()", the best source of information is the "ROM Kernal Manual, Devices", under console.device, and the autodocs, which are available on the developers CD. There is also a file on aminet with most of the information in it "aminet:dev/misc/AmigaANSI.txt" Tak Tang Experience is directly proportional to the value of equipment destroyed. ********************************************************************** This email and any files transmitted with it are confidential and intended solely for the use of the individual or entity to whom they are addressed. If you have received this email in error please notify the system manager. This footnote also confirms that this email message has been swept by MIMEsweeper for the presence of computer viruses. ********************************************************************** ------------------------------------------------------------------------ GET A NEXTCARD VISA, in 30 seconds! Get rates as low as 0.0% Intro or 9.9% Fixed APR and no hidden fees. Apply NOW! http://click.egroups.com/1/937/1/_/451227/_/953644956/ ------------------------------------------------------------------------ @ENDNODE @NODE "Mail_282" " Re: ami dev cd probs " From: Matthew Fletcher Date: 03-21-00 15:14:55 Subject: @{b}Re: ami dev cd probs@{ub} From: "Matthew Fletcher" > I have to access my cds through a samba filesharing over my tcp/ip > network from my windows pc. Windows can show the XXX.¶ files when > browsing the cd, but it cant copy them :/ > so i've managed to copy/install everything, except for the .¶ files. Easy just zip them,... or tgz or lha, or zip the whole stormc dir and then remove all but the .¶ files. -matthew ------------------------------------------------------------------------ PERFORM CPR ON YOUR APR! Get a NextCard Visa, in 30 seconds! Get rates as low as 0.0% Intro or 9.9% Fixed APR and no hidden fees. Apply NOW! http://click.egroups.com/1/2121/1/_/451227/_/953648105/ ------------------------------------------------------------------------ @ENDNODE @NODE "Mail_283" " Re: What compiler.... (Specificly, what IDE...) " From: Matthew Fletcher Date: 03-21-00 15:27:17 Subject: @{b}Re: What compiler.... (Specificly, what IDE...)@{ub} From: "Matthew Fletcher" > From: "Dominic Clifton" I got fustrated an posted to an old thread,.. so here goes. > These days IDE's on PC's have built in GUI editors, tabcompletion of > function names, excellent large project management, function lookup, > on-line context sensitive help (of the IDE and the > C-functions/keywords/APIs etc) they're just downright brilliant to > work in, everything's at your fingertips. Yes but they slow down the whole process of development, because todo tab completion you have to cache ALL fuction names into memory, constantly scan the source for what you type and generaly take up about 80% of the cpu time (on a wintel box). I can say this because after almost physicly distroying Borlands, JBuilder 2/3 and CBuilder 4 CDs, cos the bloody IDE is sooow sloooow. Just for note i have a 300mhz PII, with 80mb ram (winNT4), infront of me, and CBuilder compiles `helloworld` slower than vbcc on my 30/50 at home !!! > If I had the tools on my Amiga that the PC world enjoys today I'm > sure I'd still be coding on it as much as I used to, and I'd be > able to spend more time coding than trying to get the damn > development programs working! Err, ? what, im not a touch typest, but i can keep ahead of JBuilder 3 by a few seconds. > I've tried every amiga compiler that I've found > and I've never come across anything as easy to develop in than as > the old DOS based Turbo C was. That was great and fast to (i used the Pascal one back then) - matthew ------------------------------------------------------------------------ MAXIMIZE YOUR CARD, MINIMIZE YOUR RATE! Get a NextCard Visa, in 30 seconds! Get rates as low as 0.0% Intro or 9.9% Fixed APR and no hidden fees. Apply NOW! http://click.egroups.com/1/2122/1/_/451227/_/953648855/ ------------------------------------------------------------------------ @ENDNODE @NODE "Mail_284" " Rendering pictures correctly with MUI " From: "G. Burdett" Date: 03-21-00 16:53:50 Subject: @{b}Rendering pictures correctly with MUI@{ub} From: "G. Burdett" Hi, I got MUI working with vbcc, and now I think I've worked out how to load a byte run compressed ILBM with a BodychunkObject. However, no matter what I do, I can't get MUI to pick a decent set of colours to display my picture, even when I'm running an 8bit workbench. I tried to define an array of ULONG to describe the palette, and use the Bitmap_SourceColours and Bitmap_Precision tags, but instead of improving the pen allocation, I got a window with a big, black rectangle in it! :-( What am I doing wrong? The picture I'm trying to display has a 32 colour palette, and only a quarter of those colours are used for the actual picture (the rest of the palette will be used by other brushes - this is all for a game). My MUI code is given below... Cheers, Giles app = ApplicationObject, MUIA_Application_Title , "Roll Your Own", MUIA_Application_Version , "$VER: Roll Your Own 0.1a (20.03.00)", MUIA_Application_Copyright , "©2000, Giles Burdett", MUIA_Application_Author , "Giles Burdett", MUIA_Application_Description, "An attempt at making a MUI GUI", MUIA_Application_Base , "ROLLYOUROWN", SubWindow, window = WindowObject, MUIA_Window_Title, "Rolling Your Own", MUIA_Window_ID , MAKE_ID('R','O','L','L'), MUIA_Window_Width , MUIV_Window_Width_Screen(50), MUIA_Window_Height, MUIV_Window_Height_Screen(50), WindowContents, board = BodychunkObject, MUIA_FixWidth, 446, MUIA_FixHeight, 324, MUIA_Bitmap_Width, 446, MUIA_Bitmap_Height, 324, MUIA_Bodychunk_Depth, 6, MUIA_Bodychunk_Compression, 1, //MUIA_Bitmap_SourceColors, (ULONG *)abouttheboard, MUIA_Bodychunk_Body, theboardpic, //MUIA_Bitmap_Precision, PRECISION_GUI, End, End, End; // NB. "theboardpic" is a UBYTE pointer to the BODY chunk of // the picture. It is set to point past 'B', 'O', 'D', 'Y' // and the 32 bit integer that follows it. ------------------------------------------------------------------------ DON'T HATE YOUR RATE! Get a NextCard Visa, in 30 seconds! Get rates as low as 0.0% Intro or 9.9% Fixed APR and no hidden fees. Apply NOW! http://click.egroups.com/1/2120/1/_/451227/_/953657741/ ------------------------------------------------------------------------ @ENDNODE @NODE "Mail_285" " (Of Topic) Freeserve DNS for Amiga " From: roderick_maclean@agilent.com Date: 03-21-00 17:25:13 Subject: @{b}(Of Topic) Freeserve DNS for Amiga@{ub} From: roderick_maclean@agilent.com Apologies for posting off topic but I hoped some of you people could help me. My hard drive has died, deleting some (maybe all) of my C code but thankfully I backed it all up!!! Something more important to me has been lost though, my ISP settings. I've got a Freeserve account and password and I know the telephone number needed to dial in. I'm also pretty sure that I know the pop3 server, smtp server and news server names. If anyone could tell me the DNS server numbers (for at least one DNS server) and the name of the web-cache I would be most grateful. These details aren't widely available because it is all set up automagically on the PC but not on the lovely Amiga! I hope someone can help! Rod -- Rod MacLean | Roderick_MacLean@agilent.com Developer - Agilent Technologies | +44 (0131) 331-7304 ** My views do not necessarily reflect the views of Agilent! ** ------------------------------------------------------------------------ GET A NEXTCARD VISA, in 30 seconds! Get rates as low as 0.0% Intro or 9.9% Fixed APR and no hidden fees. Apply NOW! http://click.egroups.com/1/937/1/_/451227/_/953655929/ ------------------------------------------------------------------------ @ENDNODE @NODE "Mail_286" " Re: (Of Topic) Freeserve DNS for Amiga " From: David McMinn Date: 03-21-00 17:35:00 Subject: @{b}Re: (Of Topic) Freeserve DNS for Amiga@{ub} From: "David McMinn" Hi Roderick > I've got a Freeserve account and password and I know > the telephone number needed to dial in. 0845 0796699 > If anyone could tell me the DNS server numbers > (for at least one DNS server) 195.92.195.94 195.92.195.95 > and the name of the web-cache I would be most > grateful. http://www-cache.freeserve.net:8080 for both http and ftp > These details aren't widely available because They're on one of the FAQs on the Freeperve homepage. http://195.92.178.4/help/q_and_a.htm -- |) /\ \/ ][ |) |\/| c |\/| ][ |\| |\| | dave@satanicdreams.com http://members.xoom.com/David_McMinn | ICQ=16827694 I'd explain it to you, but your brain would explode. ------------------------------------------------------------------------ GET A NEXTCARD VISA, in 30 seconds! Get rates as low as 0.0% Intro or 9.9% Fixed APR and no hidden fees. Apply NOW! http://click.egroups.com/1/937/1/_/451227/_/953656532/ ------------------------------------------------------------------------ @ENDNODE @NODE "Mail_287" " RE: (Of Topic) Freeserve DNS for Amiga " From: "Sealey, M." Date: 03-21-00 17:44:21 Subject: @{b}RE: (Of Topic) Freeserve DNS for Amiga@{ub} From: "Sealey, M." > > If anyone could tell me the DNS server numbers > > (for at least one DNS server) > > 195.92.195.94 > 195.92.195.95 You should set Miami or whatever to autodetect DNS servers. Writing the values in is bound to cause troubles when they inevitably change them ;) For the paultry 1 or 2 seconds it takes to auto-discover the primary and secondary DNS servers you aren't going to be doing anything useful anyway. > They're on one of the FAQs on the Freeperve homepage. > http://195.92.178.4/help/q_and_a.htm Freeperve? :) -- Matt Sealey mws2@le.ac.uk Distributed Systems Support Computer Centre University of Leicester ------------------------------------------------------------------------ DON'T HATE YOUR RATE! Get a NextCard Visa, in 30 seconds! Get rates as low as 0.0% Intro or 9.9% Fixed APR and no hidden fees. Apply NOW! http://click.egroups.com/1/2120/1/_/451227/_/953657115/ ------------------------------------------------------------------------ @ENDNODE @NODE "Mail_288" " Whoops " From: "G. Burdett" Date: 03-21-00 18:08:41 Subject: @{b}Whoops@{ub} From: "G. Burdett" Hi, I posted some MUI code earlier, which has some mistakes. The Bodychunk_Depth should be 5 (sorry, the 6 was an experiment that crashed my machine and I didn't set it back before I posted the code), and the paletter array should be type UBYTE (not ULONG - doh!). Anyway, despite these changes, especially the UBYTE thing which vanquished the black rectangle problem, I still have dreadful (ie. completely wrong) colour remapping, whether I specify a precision or not. Any ideas? Cheers, Giles ------------------------------------------------------------------------ GET A NEXTCARD VISA, in 30 seconds! Get rates as low as 0.0% Intro APR and no hidden fees. Apply NOW! http://click.egroups.com/1/975/1/_/451227/_/953662200/ ------------------------------------------------------------------------ @ENDNODE @NODE "Mail_289" " New and some questions, " From: Joris Kempen Date: 03-21-00 18:47:18 Subject: @{b}New and some questions,@{ub} From: Joris Kempen First of all i would like to say hi to everyone on this list. I'm new here and hope to get and give some info about c programming. Probably the first beause i'm not that good in (Amiga) C. I did some ansi-c, read the C-Manual for the amiga and skimming through the RKRMs. (They are hard to read:( ) Now i have some questions: first of all i tried opening a window by setting up a NewWindow struc and calling OpenWindow. As the prog stops i close it with CloseWindow. This works but StormC2.0 gave me a warning that the window wasn't closed/freed but i was gone. Now i assumed that the mem wasn't released. I didn't what i did wrong so i used OpenWindowTags and heeee StormC didnt warn me anymore. Can anyone tell me what the problem me have been and was is better to use OpenWindow or OpenWindowTags? Next question: i'm trying to make a starfield on a 320*200 window on my workbench and want that when users clicks om closewindow the program stops. So i need a messageloop that isn't passive but active. How do i do that? So don't Wait( 1l << my_window->UserPort->mp_SigBit); Last question i need to redraw the window everytime because the stars are moving... how can easily clear the RPort and not lose the borders etc. I heard of a gimmezerozero window is that the solution? ------------------------------------------------------------------------ DON'T HATE YOUR RATE! Get a NextCard Visa, in 30 seconds! Get rates as low as 0.0% Intro or 9.9% Fixed APR and no hidden fees. Apply NOW! http://click.egroups.com/1/2120/1/_/451227/_/953664476/ ------------------------------------------------------------------------ @ENDNODE @NODE "Mail_290" " Re: What is reentrant code? " From: David McMinn Date: 03-21-00 19:22:54 Subject: @{b}Re: What is reentrant code?@{ub} From: "David McMinn" Hi Gabriele > I am just curious about that, what is exactly reentrant code? > Everybody talks about it but I haven't yet found what is it ! Its code that exists in memory once, but can be run as many instances. Like when you have made a program resident (or is it resident pure?) that program will be in memory only once, and it is this copy in memory that will get run when you run the command. So you could run the command 20 times and it will be the same code in the same memory that is being executed each time. Library functions are another example of this. -- |) /\ \/ ][ |) |\/| c |\/| ][ |\| |\| | dave@satanicdreams.com http://members.xoom.com/David_McMinn | ICQ=16827694 God loves stupid people. That's why he made so many. ------------------------------------------------------------------------ GET A NEXTCARD VISA, in 30 seconds! Get rates as low as 2.9% Intro or 9.9% Fixed APR and no hidden fees. Apply NOW! http://click.egroups.com/1/936/1/_/451227/_/953663005/ ------------------------------------------------------------------------ @ENDNODE @NODE "Mail_291" " Re: What compiler.... (Specificly, what IDE...) " From: Dominic Clifton Date: 03-21-00 19:43:54 Subject: @{b}Re: What compiler.... (Specificly, what IDE...)@{ub} From: "Dominic Clifton" > >After using recent IDE's on PC system's I'm amazed that I coped for so long with > >the tools that I've had on my Amigas. Even borland's DOS based Turbo > >Pascal/Turbo C had better IDE's than *any* Amiga IDE has ever had. > > Besides my gripes with Pascal, I really don't think that the Pascal DOS IDE is anything > special. > It's certinaly not better that all the Amiga IDEs I've used. no, not all of it is, but a lot of it is, for instance, it's very consistant, it has very good on-line help with examples, the source code debugging is excellent, and it's very easy to pick up. e.g. you don't have to refer to the on-line help because it was very intuitive. That was more my point. I like programs to be intuitive as it makes everything so much quicker to pick up and use. > >These days IDE's on PC's have built in GUI editors > I've never seen a built in GUI editor I've been happy with. what's wrong with the one in Visual C ? > The best I've seen IMO are (Amiga = CEd, Windows = Textpad, Linux = Vi) yup, textpad is very nice, but I prefer UltraEdit on the PC, but that's off-topic.. :-) And I've yet to find a good text editor on linux that matches amiga editiors, let alone PC ones. > Though if the Amiga had a much better market share then I'm sure we would see much > better IDEs. yup, I'm sure it would to, but linux has a big market share and it's not got a good IDE yet. (There's a few in the works though, maybe someone should port them..) > >Oh well, back to my PC for a bit of Quake 3 mod making in Visual Dev Studio... > >MMMMmmmmMMmmmmm... > > Visual Dev Studio. Yuck, Borlands C++ Builder is a better c development package IMO, I've not tried that yet, it's on my list of things todo, besides, id software used VisualDev Studio so I thought I better too (makes life easier...) > still I don't like programming under windows at all. yeah, it's just not the same is it, but it's a lot easier as so much is done for you already. Dom ------------------------------------------------------------------------ GET A NEXTCARD VISA, in 30 seconds! Get rates as low as 2.9% Intro or 9.9% Fixed APR and no hidden fees. Apply NOW! http://click.egroups.com/1/936/1/_/451227/_/953635845/ ------------------------------------------------------------------------ @ENDNODE @NODE "Mail_292" " RE: New and some questions, " From: Jesper Svennevid Date: 03-21-00 20:20:26 Subject: @{b}RE: New and some questions,@{ub} From: Jesper Svennevid > From: Joris Kempen > > First of all i would like to say hi to everyone on this list. > I'm new here and hope to get and give some info about c > programming. Probably the first beause i'm not that good in > (Amiga) C. I did some ansi-c, read the C-Manual for the amiga > and skimming through the RKRMs. (They are hard to read:( ) > > Now i have some questions: > > first of all i tried opening a window by setting up a > NewWindow struc and > calling OpenWindow. As the prog stops i close it with CloseWindow. > This works but StormC2.0 gave me a warning that the window wasn't > closed/freed but i was gone. Now i assumed that the mem > wasn't released. I > didn't what i did wrong so i used > OpenWindowTags and heeee StormC didnt warn me anymore. > Can anyone tell me what the problem me have been and was is > better to use > OpenWindow or OpenWindowTags? Sounds strange, if you called CloseWindow() on the window, it should not have warned you about it... Perhaps a bug in StormC, who knows? :) I really prefer OpenWindowTags() over OpenWindow(), as you can pass arguments directly(as OpenWindowTags is just a stub for OpenWindowTagList()) at runtime instead of feeding it to a structure, and it is much more extendable. It was many years since I last filled in a NewWindow-structure, these days I pass everything as tags and just set the NewWindow-pointer to NULL. > Next question: > i'm trying to make a starfield on a 320*200 window on my > workbench and want > that when users clicks om closewindow the program stops. So i need a > messageloop that isn't passive but active. > How do i do that? So don't Wait( 1l << > my_window->UserPort->mp_SigBit); A simple solution should be this: (I've tried to comment it a bit, but it still may be fuzzy :) /* loop until running goes FALSE (must be set to a non-FALSE value before entering loop) */ while( FALSE != running ) { /* poll the signal-mask to see if anything has arrived at the windows user-port */ if ( (1<<(my_window -> UserPort -> mpSigBit)) == SetSignal((1<<(my_window -> UserPort -> mpSigBit)),0L) ) { struct IntuiMessage* p_msg = NULL; /* loop through all waiting messages */ while( NULL != (p_msg = (struct IntuiMessage*) GetMsg( my_window -> UserPort ) { /* perform certain action depending on the class in the intuition-message */ switch( p_msg -> Class ) { case IDCMP_CLOSEWINDOW: /* tell program to terminate */ running = FALSE; break; /* add for more idcmp-messages... */ } /* reply message back to intuition */ ReplyMsg( (struct IntuiMessage*)p_msg ); } } /* here you perform your rendering */ my_window_rendering(my_window); /* wait for screen refresh */ WaitTOF(); } The best way would be to add another task that is put on a higher priority than the main-program, and that just keeeps Wait()-ing for window-actions. (This requires that you open the window without any IDCMP-flags, allocate the MsgPort in the thread and do a ModifyIDCMP() afterwards to activate IDCMP_CLOSEWINDOW) > Last question i need to redraw the window everytime because > the stars are > moving... how can easily clear the RPort and not lose the > borders etc. I > heard of a gimmezerozero window is that the solution? GimmeZeroZero is the cheap solution yes, as it allocates layers that keep the borders intact, but that is not at all necessary, you can check the BorderLeft,BorderTop,BorderRight and BorderBottom fields in the Window-structure, which contains all you need to keep your drawing within the borders of the window. Just use these as offsets into your rendering and you should have no trashing of the borders at all. -- // Jesper Svennevid : jesper@fxrealm.com Software Developer at FXRealm Studios http://www.fxrealm.com/ ------------------------------------------------------------------------ GET A NEXTCARD VISA, in 30 seconds! Get rates as low as 2.9% Intro or 9.9% Fixed APR and no hidden fees. Apply NOW! http://click.egroups.com/1/936/1/_/451227/_/953699896/ ------------------------------------------------------------------------ @ENDNODE @NODE "Mail_293" " C++ fstream.h problems " From: Daithi O'Cuinn Date: 03-21-00 20:36:54 Subject: @{b}C++ fstream.h problems@{ub} From: Daithi O'Cuinn Hello, I'm trying to compile a standard C++ program on my Amiga using gcc. I can compile other programs, but not this one, which has been migrated from a PC with MS Visual C++. I get the following error: /GNUINCLUDE/iostream.h: In method `fstream::~fstream()': /GNUINCLUDE/iostream.h:88: `ios::~ios()' is from private base class In file included from buildinfo.h:15, from test.cpp:16: /GNUINCLUDE/fstream.h:57: within this context Alas, my limited knowledge of C++ means I have no idea what this all means. It compiled alright on the PC. Is the problem with my code, or the includes? (My code would be the more likely ;) Can anyone help? Thanks, -- Daithi O'Cuinn docuinn@cluke.demon.co.uk ------------------------------------------------------------------------ PERFORM CPR ON YOUR APR! Get a NextCard Visa, in 30 seconds! Get rates as low as 0.0% Intro or 9.9% Fixed APR and no hidden fees. Apply NOW! http://click.egroups.com/1/2121/1/_/451227/_/953667668/ ------------------------------------------------------------------------ @ENDNODE @NODE "Mail_294" " Re: conio.h? " From: Donald W Millican Date: 03-21-00 20:44:01 Subject: @{b}Re: conio.h?@{ub} From: "Donald W Millican" This header file comes with the Borland Turbo C++ compiler on the PC and as far as I know was invented by them and is not part of the C / C++ language as such. I did once put it into StormC3.0 and try to make it work with my PC sourcecode and to a certain extent it did. As gotoxy() is defined in conio.h, I would assume it was invented by Borland and is unlikely to work on anyone else's compiler. I would be interested if somebody did get all of conio's functions working though. ------------------------------------------------------------------------ GET A NEXTCARD VISA, in 30 seconds! Get rates as low as 0.0% Intro APR and no hidden fees. Apply NOW! http://click.egroups.com/1/975/1/_/451227/_/953667845/ ------------------------------------------------------------------------ @ENDNODE @NODE "Mail_295" " GCC and AmiTCP includes. " From: Krzysztof Wolski Date: 03-21-00 20:45:29 Subject: @{b}GCC and AmiTCP includes.@{ub} From: Krzysztof Wolski Hello! I'm new at this list so I want to say hello :) Now I will present you my problem. I have a problem with gcc and AmiTCP includes. I wrote simple program which creates socket and then closes it. But it doesn't work :( gcc shows me warnings like this for every function defined in inline/socket.h: /gg/os-include/inline/socket.h: In function `socket': /gg/os-include/inline/socket.h:672: warning: initialization from incompatible pointer type After all warnings it shows something like this: /t/ccyL2l96.o(.text+0x58): undefined reference to `socket' /t/ccyL2l96.o(.text+0x72): undefined reference to `CloseSocket' collect2: ld returned 1 exit status After compilation there isn't file "a.out". Could you tell what should I do? (I tried to create new inlines but they don't work). Could someone send me AmiTCP includes which work with gcc? Greetings. -- Krzysztof Wolski Member of BlaBla - Polish Amiga Programming Group ------------------------------------------------------------------------ GET A NEXTCARD VISA, in 30 seconds! Get rates as low as 0.0% Intro or 9.9% Fixed APR and no hidden fees. Apply NOW! http://click.egroups.com/1/937/1/_/451227/_/953669827/ ------------------------------------------------------------------------ @ENDNODE @NODE "Mail_296" " Re: New and some questions, " From: Krzysztof Wolski Date: 03-21-00 21:05:48 Subject: @{b}Re: New and some questions,@{ub} From: Krzysztof Wolski On 21-Mar-00 Joris Kempen wrote in "[amiga-c] New and some questions,": > anymore. Can anyone tell me what the problem me have been and was is > better to use OpenWindow or OpenWindowTags? OpenWindow() is first AmigaOS function which let you open intuition windows. OpenWindowTags() works only with systems 2.0+. With this function you don't have to fill in all fields in struct NewWindow. You can choose which one you want to set up. Greetings. -- Krzysztof Wolski Member of BlaBla - Polish Amiga Programming Group ------------------------------------------------------------------------ GET A NEXTCARD VISA, in 30 seconds! Get rates as low as 0.0% Intro or 9.9% Fixed APR and no hidden fees. Apply NOW! http://click.egroups.com/1/937/1/_/451227/_/953670031/ ------------------------------------------------------------------------ @ENDNODE @NODE "Mail_297" " Re: New and some questions, " From: Joris Kempen Date: 03-21-00 21:29:55 Subject: @{b}Re: New and some questions,@{ub} From: Joris Kempen Hello Krzysztof On 21-Mar-00, you wrote: > > OpenWindow() is first AmigaOS function which let you open intuition > windows. OpenWindowTags() works only with systems 2.0+. With this function > you don't have to fill in all fields in struct NewWindow. You can choose > which one you want to set up. But when i fill my NewWindow struct i also don't need to fill in all fields. For example when i leave out the Min/Max Width/Height it still compiles without problems.. Or am i wrong about this? Aah i did some testing and you where right all fields must be filled.. Now i have another question: let's say i want to use the funtion OpenLibrary. It's located somewhere in the exec includes. So i thaught i needed to include exec/exec.h But when i do this StormC still gives me a warning that he doesn't recognize the function. This is with functions i use for AmigaOS. I found that the cure was: #include but what the heck does it do? It makes the preproces time really long. Can't this be done shorter? ------------------------------------------------------------------------ GET A NEXTCARD VISA, in 30 seconds! Get rates as low as 2.9% Intro or 9.9% Fixed APR and no hidden fees. Apply NOW! http://click.egroups.com/1/936/1/_/451227/_/953674274/ ------------------------------------------------------------------------ @ENDNODE @NODE "Mail_298" " Re: New and some questions, " From: Alexander Niven-Jenkins Date: 03-21-00 22:44:08 Subject: @{b}Re: New and some questions,@{ub} From: Alexander Niven-Jenkins Hiya Joris, On 21-Mar-00, you wrote: > Now i have another question: let's say i want to use the funtion > OpenLibrary. It's located somewhere in > the exec includes. So i thaught i needed to include exec/exec.h > But when i do this StormC still gives me a warning that he doesn't > recognize the function. This is with functions i use for AmigaOS. I would of thought that: #include is what you would want. > I found that the cure was: #include I guess this includes everthing, no wonder it takes some time :-) Kind regards... Alex -- Cremlin Software - http://www.cremlinsoftware.org Founder, lead programmer - ghandi@cremlinsoftware.org Production, programmer - anj@cremlinsoftware.org ------------------------------------------------------------------------ GET A NEXTCARD VISA, in 30 seconds! Get rates as low as 0.0% Intro APR and no hidden fees. Apply NOW! http://click.egroups.com/1/975/1/_/451227/_/953675077/ ------------------------------------------------------------------------ @ENDNODE @NODE "Mail_299" " Re: New and some questions, " From: Allan Odgaard Date: 03-21-00 22:46:13 Subject: @{b}Re: New and some questions,@{ub} From: Allan Odgaard On 21-Mar-00, Joris Kempen wrote: > Now i have another question: let's say i want to use the funtion > OpenLibrary. It's located somewhere in > the exec includes. So i thaught i needed to include exec/exec.h Generally you include "proto/.h", e.g. "proto/exec.h". > I found that the cure was: #include > but what the heck does it do? It makes the preproces time really long. Can't > this be done shorter? It includes a lot of files that you generally don't need... Regards Allan ------------------------------------------------------------------------ GET A NEXTCARD VISA, in 30 seconds! Get rates as low as 2.9% Intro or 9.9% Fixed APR and no hidden fees. Apply NOW! http://click.egroups.com/1/936/1/_/451227/_/953676660/ ------------------------------------------------------------------------ @ENDNODE @NODE "Mail_300" " Re: What is reentrant code? " From: Allan Odgaard Date: 03-21-00 22:54:08 Subject: @{b}Re: What is reentrant code?@{ub} From: Allan Odgaard On 21-Mar-00, David McMinn wrote: >> I am just curious about that, what is exactly reentrant code? >> Everybody talks about it but I haven't yet found what is it ! > Its code that exists in memory once, but can be run as many > instances. [...] More importantly two tasks can execute the code simultaneously, which basically means that the code should not write to any global variables, atleast not without semaphore protection. Regards Allan ------------------------------------------------------------------------ PERFORM CPR ON YOUR APR! Get a NextCard Visa, in 30 seconds! Get rates as low as 0.0% Intro or 9.9% Fixed APR and no hidden fees. Apply NOW! http://click.egroups.com/1/2121/1/_/451227/_/953676660/ ------------------------------------------------------------------------ @ENDNODE @NODE "Mail_301" " Re: Rendering pictures correctly with MUI " From: Allan Odgaard Date: 03-21-00 23:06:10 Subject: @{b}Re: Rendering pictures correctly with MUI@{ub} From: Allan Odgaard On 21-Mar-00, G. Burdett wrote: > I got MUI working with vbcc, and now I think I've worked out how to load > a byte run compressed ILBM with a BodychunkObject. However, Generally high colour images should be shown with the Dtpic.mui class, as this supports true colour etc. Btw: MUI questions are generally more likely to receive an answer if sent to the MUI list, also here at OneList.Com. > MUIA_Window_Width , MUIV_Window_Width_Screen(50), > MUIA_Window_Height, MUIV_Window_Height_Screen(50), This is quite dirty you know... ;-) > [...] > and the paletter array should be type UBYTE (not ULONG - doh!). Anyway, > despite these changes, especially the UBYTE thing which vanquished the black > rectangle problem, I still have dreadful (ie. completely wrong) colour > remapping, whether I specify a precision or not. The palette *should* be ULONGs, 3 pr. colour, e.g. here're the 8 MWB colours: const ULONG aboutlogo_colors[] = { 0xaaaaaaaa,0xaaaaaaaa,0xa0a0a0a0, 0x00000000,0x00000000,0x00000000, 0xffffffff,0xffffffff,0xffffffff, 0x66666666,0x88888888,0xbbbbbbbb, 0x99999999,0x99999999,0x99999999, 0xbbbbbbbb,0xbbbbbbbb,0xbbbbbbbb, 0xbbbbbbbb,0xaaaaaaaa,0x99999999, 0xffffffff,0xbbbbbbbb,0xaaaaaaaa }; Regards Allan ------------------------------------------------------------------------ PERFORM CPR ON YOUR APR! Get a NextCard Visa, in 30 seconds! Get rates as low as 0.0% Intro or 9.9% Fixed APR and no hidden fees. Apply NOW! http://click.egroups.com/1/2121/1/_/451227/_/953676661/ ------------------------------------------------------------------------ @ENDNODE @NODE "Mail_302" " Re: New and some questions, " From: Allan Odgaard Date: 03-21-00 23:08:11 Subject: @{b}Re: New and some questions,@{ub} From: Allan Odgaard On 21-Mar-00, Joris Kempen wrote: > First of all i would like to say hi to everyone on this list. > I'm new here and hope to get and give some info about c Welcome -- in the future you may want to choose subjects with care. As your letters are archived, and the archive is not very useful if every letter is just named "some questions" or similar -- it's also a good service for those who view subjects before downloading mails etc. > [...] OpenWindowTags and heeee StormC didnt warn me anymore. > Can anyone tell me what the problem me have been and was is better to use StormC patches various functions to perform resources tracking (if enabled) but don't put too much into this, as there are always countless ways to fool it, which is the case for your example... > OpenWindow or OpenWindowTags? I'd say OpenWindowTags() if your program is 2.0+. Tag functions are always better than struct functions, because they can easily be extended in a compatible way, and the programmer only need to supply the values he knows about, everything else will fall back to default values (which could be adjustable by the user). > [...] So i need a messageloop that isn't passive but active. > How do i do that? So don't Wait( 1l << my_window->UserPort->mp_SigBit); Either set up a timer that 'breaks' your message loop every now and then (when the starfield needs to be refreshed) or look at SetSignal() from exec.library, which allows you to query the signal mask. > Last question i need to redraw the window everytime because the stars are > moving... how can easily clear the RPort and not lose the borders etc. I RectFill(win->RPort, win->BorderLeft, win->BorderTop, ...); > heard of a gimmezerozero window is that the solution? These windows work by allocating two layers pr. window, which is very bad for the native layer system, i.e. it gives you a noticeable speed penalty... though it would work... Regards Allan ------------------------------------------------------------------------ GET A NEXTCARD VISA, in 30 seconds! Get rates as low as 2.9% Intro or 9.9% Fixed APR and no hidden fees. Apply NOW! http://click.egroups.com/1/936/1/_/451227/_/953676667/ ------------------------------------------------------------------------ @ENDNODE @NODE "Mail_303" " Re: Re: Single precision math functions " From: Duane McDonnell Date: 03-21-00 23:45:44 Subject: @{b}Re: Re: Single precision math functions@{ub} From: Duane McDonnell > Since I'm working on a 3d modeller I needed >fast functions for visualizing objects and I don't believe that an user >needs more than single precision floats for modelling objects (although >nobody knows...) so I went for single precision. Single precision should be sufficient, however I wouldn't call sin() and cos() at run-time for 3D work. 68881/2 FPUs are particularly slow. > But if double- >precision functions are fast enough (and I trust your compiler's code >for being faaast) I'll better jump to double. There are cycle differences for the 68k FPU range, I'm not sure about PPC however. >I probably simply drop my idea on single-precision funcs, it looks like >it's too difficult to implement and I have still a *lot* of work to do >on other things (amongst everything the interface). BTW the code from >Motorola is avaible to the public? I am really interested on >approximating transcendental functions with normal operations (w/o using >Taylor/McLaurrin formules). There are many ways to approximate sine/cosine without resorting to polynomials, if you're interested in some references let me know. However, tables are really the best way to go for high performance applications like 3D engines on slow FPU hardware. Precalulcate sine and cosine tables using sin() and cos() during setup, and fetch from these tables in your matrix code at run-time. Each entry in the table is a single precision IEEE float (32-bits). To optimize table indexing, devise your own integer angular format. A power of two value like 2048 or 4096 representing 360 degrees makes sense, since you get good resolution and easy range clipping. If you'd like the matrix and vector code modules from my 3D engine which demonstrates this, let me know and I'll send it to you. ------------------------------------------------------------------------ GET A NEXTCARD VISA, in 30 seconds! Get rates as low as 0.0% Intro APR and no hidden fees. Apply NOW! http://click.egroups.com/1/975/1/_/451227/_/953616371/ ------------------------------------------------------------------------ @ENDNODE @NODE "Mail_304" " Re: Rendering pictures correctly with MUI " From: "G. Burdett" Date: 03-22-00 00:05:23 Subject: @{b}Re: Rendering pictures correctly with MUI@{ub} From: "G. Burdett" Hi Allan, On 21-Mar-00, you wrote: > From: Allan Odgaard > > On 21-Mar-00, G. Burdett wrote: > >> I got MUI working with vbcc, and now I think I've worked out how to load >> a byte run compressed ILBM with a BodychunkObject. However, > > Generally high colour images should be shown with the >Dtpic.mui class, as this supports true colour etc. This is a 32 colour ILBM saved out from PPaint that's giving me all the grief. > Btw: MUI questions are generally more likely to receive an answer if sent > to the MUI list, also here at OneList.Com. Thanks, I'll check it out right now.... >> MUIA_Window_Width , MUIV_Window_Width_Screen(50), >> MUIA_Window_Height, MUIV_Window_Height_Screen(50), > > This is quite dirty you know... ;-) It's straight out of Balancing.c from the dev archive. Tell it to Stefan! ;) > >> [...] >> and the paletter array should be type UBYTE (not ULONG - doh!). Anyway, >> despite these changes, especially the UBYTE thing which vanquished the >> black rectangle problem, I still have dreadful (ie. completely wrong) >> colour remapping, whether I specify a precision or not. > > The palette *should* be ULONGs, 3 pr. colour, e.g. here're the 8 MWB > colours: > > const ULONG aboutlogo_colors[] = > { > 0xaaaaaaaa,0xaaaaaaaa,0xa0a0a0a0, > 0x00000000,0x00000000,0x00000000, > 0xffffffff,0xffffffff,0xffffffff, > 0x66666666,0x88888888,0xbbbbbbbb, > 0x99999999,0x99999999,0x99999999, > 0xbbbbbbbb,0xbbbbbbbb,0xbbbbbbbb, > 0xbbbbbbbb,0xaaaaaaaa,0x99999999, > 0xffffffff,0xbbbbbbbb,0xaaaaaaaa > }; > > Regards Allan > It works a hell of a lot better with UBYTE! =) This is obviously abuse of the system though... So why did the ULONG definition give me a lovely scene of blackness? I actually tried UBYTE because I thought that each gun took values from 0..255. So yeah, why is each colour encoded as a 32 bit value?! ------------------------------------------------------------------------ > PERFORM CPR ON YOUR APR! > Get a NextCard Visa, in 30 seconds! Get rates as low as > 0.0% Intro or 9.9% Fixed APR and no hidden fees. > Apply NOW! > http://click.egroups.com/1/2121/1/_/451227/_/953676661/ > ------------------------------------------------------------------------ > > Regards ------------------------------------------------------------------------ DON'T HATE YOUR RATE! Get a NextCard Visa, in 30 seconds! Get rates as low as 0.0% Intro or 9.9% Fixed APR and no hidden fees. Apply NOW! http://click.egroups.com/1/2120/1/_/451227/_/953683604/ ------------------------------------------------------------------------ @ENDNODE @NODE "Mail_305" " Re: New and some questions, " From: Colin Wenzel Date: 03-22-00 02:32:18 Subject: @{b}Re: New and some questions,@{ub} From: Colin Wenzel On 22-Mar-00, Joris Kempen wrote: > From: Joris Kempen > Hello Krzysztof > On 21-Mar-00, you wrote: >> >> OpenWindow() is first AmigaOS function which let you open intuition >> windows. OpenWindowTags() works only with systems 2.0+. With this >> function you don't have to fill in all fields in struct NewWindow. >> You can choose which one you want to set up. > > But when i fill my NewWindow struct i also don't need to fill in all > fields. > For example > when i leave out the Min/Max Width/Height it still compiles without > problems.. > > Or am i wrong about this? Absolutely, you MUST initialize all fields of the structure, even if they are zero, you must ensure that they are.... ------------------------------------- Colin Wenzel. Australia. EMAIL: mailto:colstv@hotkey.net.au URL: http://www.hotkey.net.au/~colstv/ ICQ: 17608330 AMIGA: 4000T, 060/50, 150Mb RAM, O.S 3.5 --------------------------------------- ------------------------------------------------------------------------ GET A NEXTCARD VISA, in 30 seconds! Get rates as low as 2.9% Intro or 9.9% Fixed APR and no hidden fees. Apply NOW! http://click.egroups.com/1/936/1/_/451227/_/953689588/ ------------------------------------------------------------------------ @ENDNODE @NODE "Mail_306" " Re: What is reentrant code? " From: Colin Wenzel Date: 03-22-00 02:44:49 Subject: @{b}Re: What is reentrant code?@{ub} From: Colin Wenzel On 21-Mar-00, Gabriele Svelto wrote: > From: Gabriele Svelto > > Hi everybody, > I am just curious about that, what is exactly reentrant code? > Everybody talks about it but I haven't yet found what is it ! > See ya > > Gabriele > > Basically it is a program that can be made resident so that only one instance of it exists in memory & can be used by multiple process' at the same time. The only prerequisite that I am aware of is that this program MUST dynamically allocate ALL it's writeable memory. (Via AllocMem() ) so that every copy that is started will have it's own local copy of it's writable memory. Only things that DO NOT CHANGE can be left in the program such as "dos.library", as a string constant it is not modified by the program itself, but STRUCT Library *DOSBase; MUST be allocated dynamically, because each one will want to write to 'DOSbase' variable each time one starts.. The usual way is to make a local structure of your housekeeping variable & library bases... eg: STRUCT MyJunk { STRUCT Library *DOSBase; UBYTE *MyStack; LONG MyStackLen; STRUCT Library *IntuitionBase; STRUCT Resource *ResourceBase; LONG Junk[10]; }; So, allocate a sort of STRUCT MyJunk on startup & keep your output data there.... There is one more "Gotcha", you can't keep the pointer to your memory allocation "MyJunk" in the programs data segment..... You have to make it an AUTO variable too... Don't forget to link in the right startup module too.... Also, all your linker modules must be compatible with this memory model too... Either way, it is a pain in the arse, just forget it !!! ------------------------------------- Colin Wenzel. Australia. EMAIL: mailto:colstv@hotkey.net.au URL: http://www.hotkey.net.au/~colstv/ ICQ: 17608330 AMIGA: 4000T, 060/50, 150Mb RAM, O.S 3.5 --------------------------------------- ------------------------------------------------------------------------ MAXIMIZE YOUR CARD, MINIMIZE YOUR RATE! Get a NextCard Visa, in 30 seconds! Get rates as low as 0.0% Intro or 9.9% Fixed APR and no hidden fees. Apply NOW! http://click.egroups.com/1/2122/1/_/451227/_/953689588/ ------------------------------------------------------------------------ @ENDNODE @NODE "Mail_307" " Re: Rendering pictures correctly with MUI " From: Allan Odgaard Date: 03-22-00 03:07:32 Subject: @{b}Re: Rendering pictures correctly with MUI@{ub} From: Allan Odgaard On 22-Mar-00, G. Burdett wrote: >> Generally high colour images should be shown with the >> Dtpic.mui class, as this supports true colour etc. > This is a 32 colour ILBM saved out from PPaint that's giving me all the > grief. I'd still recommend the Dtpic.mui class -- it'll probably remap faster, since Bitmap.mui uses blitter operations to remap, which is only an advantage on low colour AGA screens, and supporting high/true colour screenmodes is also important for images with less than 256 colours, as these won't need to allocate screen pens when they are on high colour screens, which means better quality and more free pens left to those programs which actually do need to allocate pens... >>> MUIA_Window_Width , MUIV_Window_Width_Screen(50), >>> MUIA_Window_Height, MUIV_Window_Height_Screen(50), >> This is quite dirty you know... ;-) > It's straight out of Balancing.c from the dev archive. Tell it to Stefan! ;) Well, if you read the autodoc entry for the two tags you'll find this comment: »Positioning and sizing should be completely under user« »control, a programmer doesn't need to worry about it.« When Stefan use them in the balance-example then it's likely because he wants to ensure the window is large enough to actually allow movement of the balance gadgets. >> The palette *should* be ULONGs, 3 pr. colour, e.g. here're the 8 MWB >> colours: [...] > It works a hell of a lot better with UBYTE! =) This is obviously abuse of > the system though... So why did the ULONG definition give me a lovely scene > of blackness? I can't say without first seeing your palette definition. Maybe the palette wasn't declared as static? > I actually tried UBYTE because I thought that each gun took values from > 0..255. So yeah, why is each colour encoded as a 32 bit value?! This is how most of graphics.library works, 32 bit offers a large precision and the values are easily scaled to the resolution of the render device, which isn't always just 0-255. Regards Allan -- URL: http://www.diku.dk/students/duff/ I am pentium of Borg. Division is futile. You'll be approximated. ------------------------------------------------------------------------ GET A NEXTCARD VISA, in 30 seconds! Get rates as low as 0.0% Intro APR and no hidden fees. Apply NOW! http://click.egroups.com/1/975/1/_/451227/_/953691094/ ------------------------------------------------------------------------ @ENDNODE @NODE "Mail_308" " Re: Opening libraries... " From: Allan Odgaard Date: 03-22-00 03:08:22 Subject: @{b}Re: Opening libraries...@{ub} From: Allan Odgaard On 21-Mar-00, Bart King wrote: >> struct Library *UtilityBase = NULL; >> if( UtilityBase = OpenLibrary( UTILITYNAME, 39 ) ) > It's a problem with StormC's includes/link libraries. As said earlier, the pasted example works fine here using StormC! Regards Allan -- URL: http://www.diku.dk/students/duff/ I have made this letter longer than usual because I lack the time to make it shorter. -- Blaise Pascal ------------------------------------------------------------------------ GET A NEXTCARD VISA, in 30 seconds! Get rates as low as 2.9% Intro or 9.9% Fixed APR and no hidden fees. Apply NOW! http://click.egroups.com/1/936/1/_/451227/_/953691095/ ------------------------------------------------------------------------ @ENDNODE @NODE "Mail_309" " Re: Rendering pictures correctly with MUI " From: "G. Burdett" Date: 03-22-00 03:25:45 Subject: @{b}Re: Rendering pictures correctly with MUI@{ub} From: "G. Burdett" Hello Allan On 22-Mar-00, you wrote: > I'd still recommend the Dtpic.mui class -- it'll probably remap faster, > since Bitmap.mui uses blitter operations to remap, which is only an > advantage on low colour AGA screens, and supporting high/true colour > screenmodes is also important for images with less than 256 colours, as > these won't need to allocate screen pens when they are on high colour > screens, which means better quality and more free pens left to those > programs which actually do need to allocate pens... Well it certainly sounds as though that method is far more system friendly. However, before I delve in to the world of the datatype class, I can't even work out how to use Bitmap, which is why I bottled out and used Bodychunk instead. All the ease of the ILBM importing, without the headache of the bitmap struct (which I've never used before). I suppose I should confess - I've been programming MUI for approximately 2 days. How am I doing?! =) (Answer: terrible :P ) > > Well, if you read the autodoc entry for the two tags you'll find this > comment: > Positioning and sizing should be completely under user > control, a programmer doesn't need to worry about it. When Stefan use > them in the balance-example then it's likely because he wants to ensure > the window is large enough to actually allow movement of the balance > gadgets. Would that mean that (by fluke, admittedly) it would be the right thing to use to stop users shrinking the window below a certain size eg. of the board of a board game? BTW, I had a look at the MUI list and noticed a thread about the MUI developer documentation. As a new, totally wet behind the ears developer (in more ways than MUI), I have to say that IMHO, the MUI documentation is a seriously tough nut to crack. After all these years, nobody has thought of a single, commonly recurring question that would find a home in a FAQ? I don't believe it! Questions like, "Why do my images look ****ing awful when I load them? I thought MUI did automatic bestpen allocation?" must have been asked by others before? Surely? Psi.c for a tutorial???!!! (Hope your listening Stefan - it's got a learning curve like climbing up a greasy lamp post - vertical and almost impossible to scale without a leg up from someone who's climbed it 1000 times before) >>> The palette *should* be ULONGs, 3 pr. colour, e.g. here're the 8 MWB >>> colours: [...] >> It works a hell of a lot better with UBYTE! =) This is obviously abuse of >> the system though... So why did the ULONG definition give me a lovely >> scene of blackness? > > I can't say without first seeing your palette definition. Maybe the > palette wasn't declared as static? I tried this and I got the black beast of MUI in my window: static const ULONG abouttheboard[96]= { 255, 255, 255, 0, 0, 0, 204, 8, 0, 235, 3, 0, 255, 42, 0, 255, 68, 0, 255, 82, 57, 247, 255, 0, 249, 224, 0, 255, 207, 37, 255, 255, 108, 255, 255, 184, 255, 255, 255, 0, 68, 136, 255, 204, 198, 255, 227, 221, 79, 57, 196, 71, 79, 170, 62, 105, 190, 74, 51, 176, 11, 0, 170, 255, 0, 204, 187, 119, 0, 153, 85, 0, 255, 170, 85, 221, 136, 68, 153, 187, 153, 102, 136, 102, 68, 170, 119, 17, 119, 85, 51, 51, 51, 238, 238, 238 }; These values entered by hand from the PPaint palette requester... I also tried writing 255L, 0L etc. but that made no difference :( What would the static do? It's a global variable but would MUI try to alter it somehow? >> I actually tried UBYTE because I thought that each gun took values from >> 0..255. So yeah, why is each colour encoded as a 32 bit value?! > > This is how most of graphics.library works, 32 bit offers a large > precision and the values are easily scaled to the resolution of the render > device, which isn't always just 0-255. I see. ULONG it is then! =) Regards, Giles ------------------------------------------------------------------------ GET A NEXTCARD VISA, in 30 seconds! Get rates as low as 0.0% Intro or 9.9% Fixed APR and no hidden fees. Apply NOW! http://click.egroups.com/1/937/1/_/451227/_/953695726/ ------------------------------------------------------------------------ @ENDNODE @NODE "Mail_310" " Re: Waiting Task (was: New and some questions) " From: Fritsch Alexander Date: 03-22-00 09:58:53 Subject: @{b}Re: Waiting Task (was: New and some questions)@{ub} From: Fritsch Alexander > From: Jesper Svennevid > > > From: Joris Kempen > Snip > > Next question: > > i'm trying to make a starfield on a 320*200 window on my > > workbench and want > > that when users clicks om closewindow the program stops. So i need a > > messageloop that isn't passive but active. > > How do i do that? So don't Wait( 1l << > > my_window->UserPort->mp_SigBit); > Snip first solution > The best way would be to add another task that is put on a higher > priority than the main-program, and that just keeeps Wait()-ing > for window-actions. (This requires that you open the window without > any IDCMP-flags, allocate the MsgPort in the thread and do a ModifyIDCMP() > afterwards to activate IDCMP_CLOSEWINDOW) > > > // Jesper Svennevid : jesper@fxrealm.com > Software Developer at FXRealm Studios > http://www.fxrealm.com/ > The idea with the new task sounds interesting. Could you please explain this a little bit further? Does that mean you dont have to poll the messages all the time? If for example I have a programm with must run as fast as possible (an emulator for instance) I dont want to check for CLOSEWINDOW or pressed Keys or Timerticks all the time. Currently I have a loop were I check the messages as you described in your example once an then call my function 100 times, but I think you Waiting-Task-idea is better. Can I use such a task for this purpose? Can you give me an example? (I have never created a task before) Kind regards Alexander Fritsch ------------------------------------------------------------------------ GET A NEXTCARD VISA, in 30 seconds! Get rates as low as 0.0% Intro APR and no hidden fees. Apply NOW! http://click.egroups.com/1/975/1/_/451227/_/953715478/ ------------------------------------------------------------------------ @ENDNODE @NODE "Mail_311" " Re: Implicit mem allocation (was: Inline) " From: Allan Odgaard Date: 03-22-00 13:36:40 Subject: @{b}Re: Implicit mem allocation (was: Inline)@{ub} From: "Allan Odgaard" --- In amiga-c@onelist.com, Jesper Svennevid wrote: > (And IMHO, to all the "we-wont-use-asm" people out there, optimizing > routines in assembler is a must sometimes even today, as the > C-compilers on amiga have a tendency on using memory way too much.) Using memory? When does a C compiler use memory? AFAIK all except stack allocations are explicit in C... ------------------------------------------------------------------------ GET A NEXTCARD VISA, in 30 seconds! Get rates as low as 2.9% Intro or 9.9% Fixed APR and no hidden fees. Apply NOW! http://click.egroups.com/1/936/1/_/451227/_/953728605/ ------------------------------------------------------------------------ @ENDNODE @NODE "Mail_312" " Re: MUI & palette (Was: Whoops) " From: Ilkka Lehtoranta Date: 03-22-00 22:07:12 Subject: @{b}Re: MUI & palette (Was: Whoops)@{ub} From: Ilkka Lehtoranta On 21-Maa-00, G. Burdett wrote: GB> paletter array should be type UBYTE (not ULONG - doh!). Anyway, despite GB> these changes, especially the UBYTE thing which vanquished the black GB> rectangle problem, I still have dreadful (ie. completely wrong) colour GB> remapping, whether I specify a precision or not. You are doing it wrong. For example white is 255,255,255 and this is in ubytes 0xff,0xff,0xff; but when you use ulongs it must be 0xffffffff,0xffffffff,0xffffffff!! Since ulong is scaled down to 8-bit 255 (0x000000ff!!) is rounded to 0x00 -> black. ------------------------------------------------------------------------ GET A NEXTCARD VISA, in 30 seconds! Get rates as low as 0.0% Intro or 9.9% Fixed APR and no hidden fees. Apply NOW! http://click.egroups.com/1/937/1/_/451227/_/953759827/ ------------------------------------------------------------------------ @ENDNODE @NODE "Mail_313" " Re: Rendering pictures correctly with MUI " From: Allan Odgaard Date: 03-22-00 22:34:45 Subject: @{b}Re: Rendering pictures correctly with MUI@{ub} From: Allan Odgaard Hej Allan, >> Positioning and sizing should be completely under user >> control, a programmer doesn't need to worry about it. > Would that mean that (by fluke, admittedly) it would be the right thing to > use to stop users shrinking the window below a certain size eg. of the > board of a board game? No, the board of a board game should be made as a class, which can report minimum dimensions, and these are always respected. So you *never* need to fiddle with the window size, as MUI will always calculate correct min, max and default sizes by looking at the contents of your window. > BTW, I had a look at the MUI list and noticed a thread about the MUI > developer documentation. As a new, totally wet behind the ears developer > (in more ways than MUI), I have to say that IMHO, the MUI documentation is > a seriously tough nut to crack. After all these years, nobody has thought > of a single, commonly recurring question that would find a home in a FAQ? I I have on a few occasions been very motivated to write a FAQ or tutorial, but it takes time and there's the mailinglist... as far as I'm concerned only one or two questions are constantly repeated, which is about eventhandlers -- but here people should just RTFM... ;-) > don't believe it! Questions like, "Why do my images look ****ing awful when > I load them? I thought MUI did automatic bestpen allocation?" must have > been asked by others before? Surely? I've been subscribed to the list for many years now, and I haven't seen it before, sorry! :-) > Psi.c for a tutorial???!!! (Hope your listening Stefan - it's got a learning > curve like climbing up a greasy lamp post - vertical and almost impossible > to scale without a leg up from someone who's climbed it 1000 times before) He isn't, but he is listening to the MUI@OneList.Com list -- I have never really looked at the PSI source, but I think it's more an example of proper use of (boopsi) classes rather than a real example of MUI programming. >> I can't say without first seeing your palette definition. Maybe the >> palette wasn't declared as static? > I tried this and I got the black beast of MUI in my window: > [...] 255, 255, 255, Okay, these values should be "expanded", as each colour is to be seen as a 32 bit left justified fraction. See the answer to the MUI list for details... > These values entered by hand from the PPaint palette requester... I think PPaint can save images as C sources that include a nice palette structure, which is 32 bit :-) > What would the static do? It's a global variable but would MUI try to > alter it somehow? No, it was just if you had it inside the scope of some procedure... Regards Allan ------------------------------------------------------------------------ MAXIMIZE YOUR CARD, MINIMIZE YOUR RATE! Get a NextCard Visa, in 30 seconds! Get rates as low as 0.0% Intro or 9.9% Fixed APR and no hidden fees. Apply NOW! http://click.egroups.com/1/2122/1/_/451227/_/953770134/ ------------------------------------------------------------------------ @ENDNODE @NODE "Mail_314" " Re: FindToolType - Illegal argument... " From: Allan Odgaard Date: 03-22-00 23:27:13 Subject: @{b}Re: FindToolType - Illegal argument...@{ub} From: Allan Odgaard On 15-Mar-00, Christian Hattemer wrote: >> I've changed the typedef for STRPTR to be 'unsigned char *' which is quite >> important when strings are used together with locale library [...] > Both versions define STRPTR like this: [...] > Why is this diffenence made for C and C++? I'm not sure, but often 'char' is not treated as signed in C++ mode, so here a string truly is just a 'char *'. Regards Allan ------------------------------------------------------------------------ GET A NEXTCARD VISA, in 30 seconds! Get rates as low as 0.0% Intro APR and no hidden fees. Apply NOW! http://click.egroups.com/1/975/1/_/451227/_/953770033/ ------------------------------------------------------------------------ @ENDNODE @NODE "Mail_315" " Re: What compiler.... " From: Allan Odgaard Date: 03-22-00 23:29:41 Subject: @{b}Re: What compiler....@{ub} From: Allan Odgaard On 15-Mar-00, Hynek Schlawack wrote: >> H&P doing AmigaOS has unfortunately stolen nearly all the time from >> StormC [...] > Sorry, but I would say, that the main work has been done by Olaf and H&P has > only coordinated the whole project. But I may be wrong however. Jochen Becher has done a lot of the OS 3.5 components, like all the preference programs, the icon editor, HDToolsBox etc. And he also seems to be the main coder on AmigaWriter, I'm not entirely sure of his role when it comes to StormC, but I know he did the editor and I think he did StormShell... Regards Allan ------------------------------------------------------------------------ GET A NEXTCARD VISA, in 30 seconds! Get rates as low as 0.0% Intro APR and no hidden fees. Apply NOW! http://click.egroups.com/1/975/1/_/451227/_/953770010/ ------------------------------------------------------------------------ @ENDNODE @NODE "Mail_316" " Re: About future compatibility... " From: Allan Odgaard Date: 03-22-00 23:31:35 Subject: @{b}Re: About future compatibility...@{ub} From: Allan Odgaard On 13-Mar-00, Gabriele Svelto wrote: > [...] Anybody thinks that those things could be removed > (if we get a new version ;-)? Thanx everybody in advance No, even if we did get a new OS version then nothing of what's currently in the autodocs/includes will be removed. Regards Allan ------------------------------------------------------------------------ GET A NEXTCARD VISA, in 30 seconds! Get rates as low as 0.0% Intro APR and no hidden fees. Apply NOW! http://click.egroups.com/1/975/1/_/451227/_/953770036/ ------------------------------------------------------------------------ @ENDNODE @NODE "Mail_317" " Re: StormC and 68000... " From: Allan Odgaard Date: 03-22-00 23:33:25 Subject: @{b}Re: StormC and 68000...@{ub} From: Allan Odgaard On 16-Mar-00, Colin Wenzel wrote: >>> Do think it makes sence today to support the MC68000? [...] >> I always defaults to 68020 and kick 3.0 -- [...] > But then technically it wouldn't be a 68xxx compiler anymore. I don't really understand this argument. > Plus !!! What am I going to do with my A600 ??? If you use it, then stick to 68000. Regards Allan -- URL: http://www.diku.dk/students/duff/ Windows - Microsoft's most popular DOS program. ------------------------------------------------------------------------ MAXIMIZE YOUR CARD, MINIMIZE YOUR RATE! Get a NextCard Visa, in 30 seconds! Get rates as low as 0.0% Intro or 9.9% Fixed APR and no hidden fees. Apply NOW! http://click.egroups.com/1/2122/1/_/451227/_/953770036/ ------------------------------------------------------------------------ @ENDNODE @NODE "Mail_318" " Re: Re: Opening libraries... " From: Bart King Date: 03-23-00 00:07:08 Subject: @{b}Re: Re: Opening libraries...@{ub} From: Bart King Allan Odgaard wrote: > As said earlier, the pasted example works fine here using StormC! Heh. Yeah, the pasted example works here too, but opening utility.library in my own program didn't. I reckon some includes or something screw it around somewhere. Nothing wrong with auto-opening utility.library though. -- Bart King Developer http://www.bartman.demon.co.uk (not up yet - soon) http://www.fscked.net/fubar/ ICQ: 18178781 ------------------------------------------------------------------------ GET A NEXTCARD VISA, in 30 seconds! Get rates as low as 2.9% Intro or 9.9% Fixed APR and no hidden fees. Apply NOW! http://click.egroups.com/1/936/1/_/451227/_/953766550/ ------------------------------------------------------------------------ @ENDNODE @NODE "Mail_319" " Re: Inline " From: Bart King Date: 03-23-00 00:09:44 Subject: @{b}Re: Inline@{ub} From: Bart King Jesper Svennevid wrote: > The only amiga C-compilers I know of that supports inlining assembler-code > is GCC(like you said), and VBCC. Damn it, I must really try using VBCC :) I know it's on Aminet, but er... I can't access any of the mirrors for some reason at the moment. D'oh. -- Bart King Developer http://www.bartman.demon.co.uk (not up yet - soon) http://www.fscked.net/fubar/ ICQ: 18178781 ------------------------------------------------------------------------ MAXIMIZE YOUR CARD, MINIMIZE YOUR RATE! Get a NextCard Visa, in 30 seconds! Get rates as low as 0.0% Intro or 9.9% Fixed APR and no hidden fees. Apply NOW! http://click.egroups.com/1/2122/1/_/451227/_/953766878/ ------------------------------------------------------------------------ @ENDNODE @NODE "Mail_320" " Re: MUI & palette (Was: Whoops) " From: "G. Burdett" Date: 03-23-00 01:01:38 Subject: @{b}Re: MUI & palette (Was: Whoops)@{ub} From: "G. Burdett" Hi Ilkka, On 22-Mar-00, you wrote: > From: Ilkka Lehtoranta > You are doing it wrong. For example white is 255,255,255 and this > is in ubytes 0xff,0xff,0xff; but when you use ulongs it must be > 0xffffffff,0xffffffff,0xffffffff!! Since ulong is scaled down to 8-bit > 255 (0x000000ff!!) is rounded to 0x00 -> black. I see you what you mean. I thought it just had to be 32 bits wide, but I didn't realise the colour information had to be the MSBs. Saved again! :) Cheers, Giles ------------------------------------------------------------------------ GET A NEXTCARD VISA, in 30 seconds! Get rates as low as 0.0% Intro APR and no hidden fees. Apply NOW! http://click.egroups.com/1/975/1/_/451227/_/953775211/ ------------------------------------------------------------------------ @ENDNODE @NODE "Mail_321" " Re: StormC and 68000... " From: Colin Wenzel Date: 03-23-00 03:47:16 Subject: @{b}Re: StormC and 68000...@{ub} From: Colin Wenzel On 23-Mar-00, Allan Odgaard wrote: > From: Allan Odgaard > On 16-Mar-00, Colin Wenzel wrote: >>>> Do think it makes sence today to support the MC68000? [...] >>> I always defaults to 68020 and kick 3.0 -- [...] >> But then technically it wouldn't be a 68xxx compiler anymore. > > I don't really understand this argument. I am being pedantic, what I was saying is that if you say you have a 68000 compiler that doesn't actually compile code for a 68000, then technically, it isn't a 68000 compiler.... (:> >> Plus !!! What am I going to do with my A600 ??? > If you use it, then stick to 68000. Who said Germans don't have a sence of humor. (:/ ------------------------------------- Colin Wenzel. Australia. EMAIL: mailto:colstv@hotkey.net.au URL: http://www.hotkey.net.au/~colstv/ ICQ: 17608330 AMIGA: 4000T, 060/50, 150Mb RAM, O.S 3.5 --------------------------------------- ------------------------------------------------------------------------ PERFORM CPR ON YOUR APR! Get a NextCard Visa, in 30 seconds! Get rates as low as 0.0% Intro or 9.9% Fixed APR and no hidden fees. Apply NOW! http://click.egroups.com/1/2121/1/_/451227/_/953779782/ ------------------------------------------------------------------------ @ENDNODE @NODE "Mail_322" " Check this " From: BOYER Laurent Date: 03-23-00 10:37:37 Subject: @{b}Check this@{ub} Have fun with these links. Bye. @ENDNODE @NODE "Mail_323" " Watch out BOYER Laurent " From: "Andrzej J. Debicki" Date: 03-23-00 11:35:31 Subject: @{b}Watch out BOYER Laurent @{ub} From: "Andrzej J. Debicki" Hello, He's sending VBS/FreeLink virus (for peecee)!!! Delete email with subject: "[amiga-c] Check this" with attachment: "links.vbs" -- _ # Andrzej J. Debicki # A12ooT 66MB \\ _ # PANS - Polish Amiga News Squad # 6o3e/o6o \\// # denbi@power-amiga.prv.pl # BVisionPPC \/ # http://www.power-amiga.prv.pl # AmigaOS 3.5 ----------------------------------------------------------- ------------------------------------------------------------------------ GET A NEXTCARD VISA, in 30 seconds! Get rates as low as 0.0% Intro APR and no hidden fees. Apply NOW! http://click.egroups.com/1/975/1/_/451227/_/953807742/ ------------------------------------------------------------------------ @ENDNODE @NODE "Mail_324" " RE: Re: Implicit mem allocation (was: Inline) " From: Jesper Svennevid Date: 03-23-00 11:47:49 Subject: @{b}RE: Re: Implicit mem allocation (was: Inline)@{ub} From: Jesper Svennevid > --- In amiga-c@onelist.com, Jesper Svennevid wrote: > > (And IMHO, to all the "we-wont-use-asm" people out there, optimizing > > routines in assembler is a must sometimes even today, as the > > C-compilers on amiga have a tendency on using memory way too much.) > > Using memory? When does a C compiler use memory? AFAIK all except > stack allocations are explicit in C... I mean't ACCESSING memory, variables that exist in memory-space is slower than ones exisiting in register-space, especially if they happen to be loop-counters in some tight copy-loop, or pointers that are only used to move memory from one position to another. Example: void my_copy( unsigned long* dest, unsigned long* source, size_t size ) { unsigned long* d_end = dest + (size>>2); while(dest